/modules/System/includes/XPath.class.php

Large files files are truncated, but you can click here to view the full file

<?php
/**
 * Php.XPath
 *
 * +======================================================================================================+
 * | A php class for searching an XML document using XPath, and making modifications using a DOM 
 * | style API. Does not require the DOM XML PHP library. 
 * |
 * +======================================================================================================+
 * | What Is XPath:
 * | --------------
 * | - "What SQL is for a relational database, XPath is for an XML document." -- Sam Blum
 * | - "The primary purpose of XPath is to address parts of an XML document. In support of this 
 * |    primary purpose, it also provides basic facilities for manipulting it." -- W3C
 * | 
 * | XPath in action and a very nice intro is under:
 * |    http://www.zvon.org/xxl/XPathTutorial/General/examples.html
 * | Specs Can be found under:
 * |    http://www.w3.org/TR/xpath     W3C XPath Recommendation 
 * |    http://www.w3.org/TR/xpath20   W3C XPath Recommendation 
 * |
 * | NOTE: Most of the XPath-spec has been realized, but not all. Usually this should not be
 * |       problem as the missing part is either rarely used or it's simpler to do with PHP itself.
 * +------------------------------------------------------------------------------------------------------+
 * | Requires PHP version  4.0.5 and up
 * +------------------------------------------------------------------------------------------------------+
 * | Main Active Authors:
 * | --------------------
 * | Nigel Swinson <nigelswinson@users.sourceforge.net>
 * |   Started around 2001-07, saved phpxml from near death and renamed to Php.XPath
 * |   Restructured XPath code to stay in line with XPath spec.
 * | Sam Blum <bs_php@infeer.com>
 * |   Started around 2001-09 1st major restruct (V2.0) and testbench initiator.   
 * |   2nd (V3.0) major rewrite in 2002-02
 * | Daniel Allen <bigredlinux@yahoo.com>
 * |   Started around 2001-10 working to make Php.XPath adhere to specs 
 * | Main Former Author: Michael P. Mehl <mpm@phpxml.org>
 * |   Inital creator of V 1.0. Stoped activities around 2001-03        
 * +------------------------------------------------------------------------------------------------------+
 * | Code Structure:
 * | --------------_
 * | The class is split into 3 main objects. To keep usability easy all 3 
 * | objects are in this file (but may be split in 3 file in future).
 * |   +-------------+ 
 * |   |  XPathBase  | XPathBase holds general and debugging functions. 
 * |   +------+------+
 * |          v      
 * |   +-------------+ XPathEngine is the implementation of the W3C XPath spec. It contains the 
 * |   | XPathEngine | XML-import (parser), -export  and can handle xPathQueries. It's a fully 
 * |   +------+------+ functional class but has no functions to modify the XML-document (see following).
 * |          v      
 * |   +-------------+ 
 * |   |    XPath    | XPath extends the functionality with actions to modify the XML-document.
 * |   +-------------+ We tryed to implement a DOM - like interface.
 * +------------------------------------------------------------------------------------------------------+
 * | Usage:
 * | ------
 * | Scroll to the end of this php file and you will find a short sample code to get you started
 * +------------------------------------------------------------------------------------------------------+
 * | Glossary:
 * | ---------
 * | To understand how to use the functions and to pass the right parameters, read following:
 * |     
 * | Document: (full node tree, XML-tree)
 * |     After a XML-source has been imported and parsed, it's stored as a tree of nodes sometimes 
 * |     refered to as 'document'.
 * |     
 * | AbsoluteXPath: (xPath, xPathSet)
 * |     A absolute XPath is a string. It 'points' to *one* node in the XML-document. We use the
 * |     term 'absolute' to emphasise that it is not an xPath-query (see xPathQuery). A valid xPath 
 * |     has the form like '/AAA[1]/BBB[2]/CCC[1]'. Usually functions that require a node (see Node) 
 * |     will also accept an abs. XPath.
 * |     
 * | Node: (node, nodeSet, node-tree)
 * |     Some funtions require or return a node (or a whole node-tree). Nodes are only used with the 
 * |     XPath-interface and have an internal structure. Every node in a XML document has a unique 
 * |     corresponding abs. xPath. That's why public functions that accept a node, will usually also 
 * |     accept a abs. xPath (a string) 'pointing' to an existing node (see absolutXPath).
 * |     
 * | XPathQuery: (xquery, query)
 * |     A xPath-query is a string that is matched against the XML-document. The result of the match 
 * |     is a xPathSet (vector of xPath's). It's always possible to pass a single absoluteXPath 
 * |     instead of a xPath-query. A valid xPathQuery could look like this:
 * |     '//XXX/*[contains(., "foo")]/..' (See the link in 'What Is XPath' to learn more).
 * |     
 * |     
 * +------------------------------------------------------------------------------------------------------+
 * | Internals:
 * | ----------
 * | - The Node Tree
 * |   -------------
 * | A central role of the package is how the XML-data is stored. The whole data is in a node-tree.
 * | A node can be seen as the equvalent to a tag in the XML soure with some extra info.
 * | For instance the following XML 
 * |                        <AAA foo="x">***<BBB/><CCC/>**<BBB/>*</AAA>
 * | Would produce folowing node-tree:
 * |                              'super-root'      <-- $nodeRoot (Very handy)  
 * |                                    |                                           
 * |             'depth' 0            AAA[1]        <-- top node. The 'textParts' of this node would be
 * |                                /   |   \                     'textParts' => array('***','','**','*')
 * |             'depth' 1     BBB[1] CCC[1] BBB[2]               (NOTE: Is always size of child nodes+1)
 * | - The Node
 * |   --------
 * | The node itself is an structure desiged mainly to be used in connection with the interface of PHP.XPath.
 * | That means it's possible for functions to return a sub-node-tree that can be used as input of an other 
 * | PHP.XPath function.
 * | 
 * | The main structure of a node is:
 * |   $node = array(
 * |     'name'        => '',      # The tag name. E.g. In <FOO bar="aaa"/> it would be 'FOO'
 * |     'attributes'  => array(), # The attributes of the tag E.g. In <FOO bar="aaa"/> it would be array('bar'=>'aaa')
 * |     'textParts'   => array(), # Array of text parts surrounding the children E.g. <FOO>aa<A>bb<B/>cc</A>dd</FOO> -> array('aa','bb','cc','dd')
 * |     'childNodes'  => array(), # Array of refences (pointers) to child nodes.
 * |     
 * | For optimisation reasions some additional data is stored in the node too:
 * |     'parentNode'  => NULL     # Reference (pointer) to the parent node (or NULL if it's 'super root')
 * |     'depth'       => 0,       # The tag depth (or tree level) starting with the root tag at 0.
 * |     'pos'         => 0,       # Is the zero-based position this node has in the parent's 'childNodes'-list.
 * |     'contextPos'  => 1,       # Is the one-based position this node has by counting the siblings tags (tags with same name)
 * |     'xpath'       => ''       # Is the abs. XPath to this node.
 * |     'generated_id'=> ''       # The id returned for this node by generate-id() (attribute and text nodes not supported)
 * | 
 * | - The NodeIndex
 * |   -------------
 * | Every node in the tree has an absolute XPath. E.g '/AAA[1]/BBB[2]' the $nodeIndex is a hash array
 * | to all the nodes in the node-tree. The key used is the absolute XPath (a string).
 * |    
 * +------------------------------------------------------------------------------------------------------+
 * | License:
 * | --------
 * | The contents of this file are subject to the Mozilla Public License Version 1.1 (the "License"); 
 * | you may not use this file except in compliance with the License. You may obtain a copy of the 
 * | License at http://www.mozilla.org/MPL/ 
 * | 
 * | Software distributed under the License is distributed on an "AS IS" basis, WITHOUT WARRANTY
 * | OF ANY KIND, either express or implied. See the License for the specific language governing 
 * | rights and limitations under the License. 
 * |
 * | The Original Code is <phpXML/>. 
 * | 
 * | The Initial Developer of the Original Code is Michael P. Mehl. Portions created by Michael 
 * | P. Mehl are Copyright (C) 2001 Michael P. Mehl. All Rights Reserved.
 * |
 * | Contributor(s): N.Swinson / S.Blum / D.Allen
 * | 
 * | Alternatively, the contents of this file may be used under the terms of either of the GNU 
 * | General Public License Version 2 or later (the "GPL"), or the GNU Lesser General Public 
 * | License Version 2.1 or later (the "LGPL"), in which case the provisions of the GPL or the 
 * | LGPL License are applicable instead of those above.  If you wish to allow use of your version 
 * | of this file only under the terms of the GPL or the LGPL License and not to allow others to 
 * | use your version of this file under the MPL, indicate your decision by deleting the 
 * | provisions above and replace them with the notice and other provisions required by the 
 * | GPL or the LGPL License.  If you do not delete the provisions above, a recipient may use 
 * | your version of this file under either the MPL, the GPL or the LGPL License. 
 * | 
 * +======================================================================================================+
 *
 * @author  S.Blum / N.Swinson / D.Allen / (P.Mehl)
 * @link    http://sourceforge.net/projects/phpxpath/
 * @version 3.5
 * @CVS $Id: XPath.class.php,v 1.9 2005/11/16 17:26:05 bigmichi1 Exp $
 */

// Include guard, protects file being included twice
$ConstantName = 'INCLUDED_'.strtoupper(__FILE__);
if (defined($ConstantName)) return;
define($ConstantName,1, TRUE);

/************************************************************************************************
* ===============================================================================================
*                               X P a t h B a s e  -  Class                                      
* ===============================================================================================
************************************************************************************************/
class XPathBase {
  var $_lastError;
  
  // As debugging of the xml parse is spread across several functions, we need to make this a member.
  var $bDebugXmlParse = FALSE;

  // do we want to do profiling?
  var $bClassProfiling = FALSE;

  // Used to help navigate through the begin/end debug calls
  var $iDebugNextLinkNumber = 1;
  var $aDebugOpenLinks = array();
  var $aDebugFunctions = array(
          //'_evaluatePrimaryExpr',
          //'_evaluateExpr',
          //'_evaluateStep',
          //'_checkPredicates',
          //'_evaluateFunction',
          //'_evaluateOperator',
          //'_evaluatePathExpr',
               );

  /**
   * Constructor
   */
  function XPathBase() {
    # $this->bDebugXmlParse = TRUE;
    $this->properties['verboseLevel'] = 1;  // 0=silent, 1 and above produce verbose output (an echo to screen). 
    
    if (!isSet($_ENV)) {  // Note: $_ENV introduced in 4.1.0. In earlier versions, use $HTTP_ENV_VARS.
      $_ENV = $GLOBALS['HTTP_ENV_VARS'];
    }
    
    // Windows 95/98 do not support file locking. Detecting OS (Operation System) and setting the 
    // properties['OS_supports_flock'] to FALSE if win 95/98 is detected. 
    // This will surpress the file locking error reported from win 98 users when exportToFile() is called.
    // May have to add more OS's to the list in future (Macs?).
    // ### Note that it's only the FAT and NFS file systems that are really a problem.  NTFS and
    // the latest php libs do support flock()
    $_ENV['OS'] = isSet($_ENV['OS']) ? $_ENV['OS'] : 'Unknown OS';
    switch ($_ENV['OS']) { 
      case 'Windows_95':
      case 'Windows_98':
      case 'Unknown OS':
        // should catch Mac OS X compatible environment 
        if (!empty($_SERVER['SERVER_SOFTWARE']) 
            && preg_match('/Darwin/',$_SERVER['SERVER_SOFTWARE'])) { 
           // fall-through 
        } else { 
           $this->properties['OS_supports_flock'] = FALSE; 
           break; 
        }
      default:
        $this->properties['OS_supports_flock'] = TRUE;
    }
  }
  
  
  /**
   * Resets the object so it's able to take a new xml sting/file
   *
   * Constructing objects is slow.  If you can, reuse ones that you have used already
   * by using this reset() function.
   */
  function reset() {
    $this->_lastError   = '';
  }
  
  //-----------------------------------------------------------------------------------------
  // XPathBase                    ------  Helpers  ------                                    
  //-----------------------------------------------------------------------------------------
  
  /**
   * This method checks the right amount and match of brackets
   *
   * @param     $term (string) String in which is checked.
   * @return          (bool)   TRUE: OK / FALSE: KO  
   */
  function _bracketsCheck($term) {
    $leng = strlen($term);
    $brackets = 0;
    $bracketMisscount = $bracketMissmatsh = FALSE;
    $stack = array();
    for ($i=0; $i<$leng; $i++) {
      switch ($term[$i]) {
        case '(' : 
        case '[' : 
          $stack[$brackets] = $term[$i]; 
          $brackets++; 
          break;
        case ')': 
          $brackets--;
          if ($brackets<0) {
            $bracketMisscount = TRUE;
            break 2;
          }
          if ($stack[$brackets] != '(') {
            $bracketMissmatsh = TRUE;
            break 2;
          }
          break;
        case ']' : 
          $brackets--;
          if ($brackets<0) {
            $bracketMisscount = TRUE;
            break 2;
          }
          if ($stack[$brackets] != '[') {
            $bracketMissmatsh = TRUE;
            break 2;
          }
          break;
      }
    }
    // Check whether we had a valid number of brackets.
    if ($brackets != 0) $bracketMisscount = TRUE;
    if ($bracketMisscount || $bracketMissmatsh) {
      return FALSE;
    }
    return TRUE;
  }
  
  /**
   * Looks for a string within another string -- BUT the search-string must be located *outside* of any brackets.
   *
   * This method looks for a string within another string. Brackets in the
   * string the method is looking through will be respected, which means that
   * only if the string the method is looking for is located outside of
   * brackets, the search will be successful.
   *
   * @param     $term       (string) String in which the search shall take place.
   * @param     $expression (string) String that should be searched.
   * @return                (int)    This method returns -1 if no string was found, 
   *                                 otherwise the offset at which the string was found.
   */
  function _searchString($term, $expression) {
    $bracketCounter = 0; // Record where we are in the brackets. 
    $leng = strlen($term);
    $exprLeng = strlen($expression);
    for ($i=0; $i<$leng; $i++) {
      $char = $term[$i];
      if ($char=='(' || $char=='[') {
        $bracketCounter++;
        continue;
      }
      elseif ($char==')' || $char==']') {
        $bracketCounter--;
      }
      if ($bracketCounter == 0) {
        // Check whether we can find the expression at this index.
        if (substr($term, $i, $exprLeng) == $expression) return $i;
      }
    }
    // Nothing was found.
    return (-1);
  }
  
  /**
   * Split a string by a searator-string -- BUT the separator-string must be located *outside* of any brackets.
   * 
   * Returns an array of strings, each of which is a substring of string formed 
   * by splitting it on boundaries formed by the string separator. 
   *
   * @param     $separator  (string) String that should be searched.
   * @param     $term       (string) String in which the search shall take place.
   * @return                (array)  see above
   */
  function _bracketExplode($separator, $term) {
    // Note that it doesn't make sense for $separator to itself contain (,),[ or ],
    // but as this is a private function we should be ok.
    $resultArr   = array();
    $bracketCounter = 0;  // Record where we are in the brackets. 
    do { // BEGIN try block
      // Check if any separator is in the term
      $sepLeng =  strlen($separator);
      if (strpos($term, $separator)===FALSE) { // no separator found so end now
        $resultArr[] = $term;
        break; // try-block
      }
      
      // Make a substitute separator out of 'unused chars'.
      $substituteSep = str_repeat(chr(2), $sepLeng);
      
      // Now determine the first bracket '(' or '['.
      $tmp1 = strpos($term, '(');
      $tmp2 = strpos($term, '[');
      if ($tmp1===FALSE) {
        $startAt = (int)$tmp2;
      } elseif ($tmp2===FALSE) {
        $startAt = (int)$tmp1;
      } else {
        $startAt = min($tmp1, $tmp2);
      }
      
      // Get prefix string part before the first bracket.
      $preStr = substr($term, 0, $startAt);
      // Substitute separator in prefix string.
      $preStr = str_replace($separator, $substituteSep, $preStr);
      
      // Now get the rest-string (postfix string)
      $postStr = substr($term, $startAt);
      // Go all the way through the rest-string.
      $strLeng = strlen($postStr);
      for ($i=0; $i < $strLeng; $i++) {
        $char = $postStr[$i];
        // Spot (,),[,] and modify our bracket counter.  Note there is an
        // assumption here that you don't have a string(with[mis)matched]brackets.
        // This should be ok as the dodgy string will be detected elsewhere.
        if ($char=='(' || $char=='[') {
          $bracketCounter++;
          continue;
        } 
        elseif ($char==')' || $char==']') {
          $bracketCounter--;
        }
        // If no brackets surround us check for separator
        if ($bracketCounter == 0) {
          // Check whether we can find the expression starting at this index.
          if ((substr($postStr, $i, $sepLeng) == $separator)) {
            // Substitute the found separator 
            for ($j=0; $j<$sepLeng; $j++) {
              $postStr[$i+$j] = $substituteSep[$j];
            }
          }
        }
      }
      // Now explod using the substitute separator as key.
      $resultArr = explode($substituteSep, $preStr . $postStr);
    } while (FALSE); // End try block
    // Return the results that we found. May be a array with 1 entry.
    return $resultArr;
  }

  /**
   * Split a string at it's groups, ie bracketed expressions
   * 
   * Returns an array of strings, when concatenated together would produce the original
   * string.  ie a(b)cde(f)(g) would map to:
   * array ('a', '(b)', cde', '(f)', '(g)')
   *
   * @param     $string  (string) The string to process
   * @param     $open    (string) The substring for the open of a group
   * @param     $close   (string) The substring for the close of a group
   * @return             (array)  The parsed string, see above
   */
  function _getEndGroups($string, $open='[', $close=']') {
    // Note that it doesn't make sense for $separator to itself contain (,),[ or ],
    // but as this is a private function we should be ok.
    $resultArr   = array();
    do { // BEGIN try block
      // Check if we have both an open and a close tag      
      if (empty($open) and empty($close)) { // no separator found so end now
        $resultArr[] = $string;
        break; // try-block
      }

      if (empty($string)) {
        $resultArr[] = $string;
        break; // try-block
      }

      
      while (!empty($string)) {
        // Now determine the first bracket '(' or '['.
        $openPos = strpos($string, $open);
        $closePos = strpos($string, $close);
        if ($openPos===FALSE || $closePos===FALSE) {
          // Oh, no more groups to be found then.  Quit
          $resultArr[] = $string;
          break;
        }

        // Sanity check
        if ($openPos > $closePos) {
          // Malformed string, dump the rest and quit.
          $resultArr[] = $string;
          break;
        }

        // Get prefix string part before the first bracket.
        $preStr = substr($string, 0, $openPos);
        // This is the first string that will go in our output
        if (!empty($preStr))
          $resultArr[] = $preStr;

        // Skip over what we've proceed, including the open char
        $string = substr($string, $openPos + 1 - strlen($string));

        // Find the next open char and adjust our close char
//echo "close: $closePos\nopen: $openPos\n\n";
        $closePos -= $openPos + 1;
        $openPos = strpos($string, $open);
//echo "close: $closePos\nopen: $openPos\n\n";

        // While we have found nesting...
        while ($openPos && $closePos && ($closePos > $openPos)) {
          // Find another close pos after the one we are looking at
          $closePos = strpos($string, $close, $closePos + 1);
          // And skip our open
          $openPos = strpos($string, $open, $openPos + 1);
        }
//echo "close: $closePos\nopen: $openPos\n\n";

        // If we now have a close pos, then it's the end of the group.
        if ($closePos === FALSE) {
          // We didn't... so bail dumping what was left
          $resultArr[] = $open.$string;
          break;
        }

        // We did, so we can extract the group
        $resultArr[] = $open.substr($string, 0, $closePos + 1);
        // Skip what we have processed
        $string = substr($string, $closePos + 1);
      }
    } while (FALSE); // End try block
    // Return the results that we found. May be a array with 1 entry.
    return $resultArr;
  }
  
  /**
   * Retrieves a substring before a delimiter.
   *
   * This method retrieves everything from a string before a given delimiter,
   * not including the delimiter.
   *
   * @param     $string     (string) String, from which the substring should be extracted.
   * @param     $delimiter  (string) String containing the delimiter to use.
   * @return                (string) Substring from the original string before the delimiter.
   * @see       _afterstr()
   */
  function _prestr(&$string, $delimiter, $offset=0) {
    // Return the substring.
    $offset = ($offset<0) ? 0 : $offset;
    $pos = strpos($string, $delimiter, $offset);
    if ($pos===FALSE) return $string; else return substr($string, 0, $pos);
  }
  
  /**
   * Retrieves a substring after a delimiter.
   *
   * This method retrieves everything from a string after a given delimiter,
   * not including the delimiter.
   *
   * @param     $string     (string) String, from which the substring should be extracted.
   * @param     $delimiter  (string) String containing the delimiter to use.
   * @return                (string) Substring from the original string after the delimiter.
   * @see       _prestr()
   */
  function _afterstr($string, $delimiter, $offset=0) {
    $offset = ($offset<0) ? 0 : $offset;
    // Return the substring.
    return substr($string, strpos($string, $delimiter, $offset) + strlen($delimiter));
  }
  
  //-----------------------------------------------------------------------------------------
  // XPathBase                ------  Debug Stuff  ------                                    
  //-----------------------------------------------------------------------------------------
  
  /**
   * Alter the verbose (error) level reporting.
   *
   * Pass an int. >0 to turn on, 0 to turn off.  The higher the number, the 
   * higher the level of verbosity. By default, the class has a verbose level 
   * of 1.
   *
   * @param $levelOfVerbosity (int) default is 1 = on
   */
  function setVerbose($levelOfVerbosity = 1) {
    $level = -1;
    if ($levelOfVerbosity === TRUE) {
      $level = 1;
    } elseif ($levelOfVerbosity === FALSE) {
      $level = 0;
    } elseif (is_numeric($levelOfVerbosity)) {
      $level = $levelOfVerbosity;
    }
    if ($level >= 0) $this->properties['verboseLevel'] = $levelOfVerbosity;
  }
   
  /**
   * Returns the last occured error message.
   *
   * @access public
   * @return string (may be empty if there was no error at all)
   * @see    _setLastError(), _lastError
   */
  function getLastError() {
    return $this->_lastError;
  }
  
  /**
   * Creates a textual error message and sets it. 
   * 
   * example: 'XPath error in THIS_FILE_NAME:LINE. Message: YOUR_MESSAGE';
   * 
   * I don't think the message should include any markup because not everyone wants to debug 
   * into the browser window.
   * 
   * You should call _displayError() rather than _setLastError() if you would like the message,
   * dependant on their verbose settings, echoed to the screen.
   * 
   * @param $message (string) a textual error message default is ''
   * @param $line    (int)    the line number where the error occured, use __LINE__
   * @see getLastError()
   */
  function _setLastError($message='', $line='-', $file='-') {
    $this->_lastError = 'XPath error in ' . basename($file) . ':' . $line . '. Message: ' . $message;
  }
  
  /**
   * Displays an error message.
   *
   * This method displays an error messages depending on the users verbose settings 
   * and sets the last error message.  
   *
   * If also possibly stops the execution of the script.
   * ### Terminate should not be allowed --fab.  Should it??  N.S.
   *
   * @param $message    (string)  Error message to be displayed.
   * @param $lineNumber (int)     line number given by __LINE__
   * @param $terminate  (bool)    (default TURE) End the execution of this script.
   */
  function _displayError($message, $lineNumber='-', $file='-', $terminate=TRUE) {
    // Display the error message.
    $err = '<b>XPath error in '.basename($file).':'.$lineNumber.'</b> '.$message."<br \>\n";
    $this->_setLastError($message, $lineNumber, $file);
    if (($this->properties['verboseLevel'] > 0) OR ($terminate)) echo $err;
    // End the execution of this script.
    if ($terminate) exit;
  }

  /**
   * Displays a diagnostic message
   *
   * This method displays an error messages
   *
   * @param $message    (string)  Error message to be displayed.
   * @param $lineNumber (int)     line number given by __LINE__
   */
  function _displayMessage($message, $lineNumber='-', $file='-') {
    // Display the error message.
    $err = '<b>XPath message from '.basename($file).':'.$lineNumber.'</b> '.$message."<br \>\n";
    if ($this->properties['verboseLevel'] > 0) echo $err;
  }
  
  /**
   * Called to begin the debug run of a function.
   *
   * This method starts a <DIV><PRE> tag so that the entry to this function
   * is clear to the debugging user.  Call _closeDebugFunction() at the
   * end of the function to create a clean box round the function call.
   *
   * @author    Nigel Swinson <nigelswinson@users.sourceforge.net>
   * @author    Sam   Blum    <bs_php@infeer.com>
   * @param     $functionName (string) the name of the function we are beginning to debug
   * @param     $bDebugFlag   (bool) TRUE if we are to draw a call stack, FALSE otherwise
   * @return                  (array)  the output from the microtime() function.
   * @see       _closeDebugFunction()
   */
  function _beginDebugFunction($functionName, $bDebugFlag) {
    if ($bDebugFlag) {
      $fileName = basename(__FILE__);
      static $color = array('green','blue','red','lime','fuchsia', 'aqua');
      static $colIndex = -1;
      $colIndex++;
      echo '<div style="clear:both" align="left"> ';
      echo '<pre STYLE="border:solid thin '. $color[$colIndex % 6] . '; padding:5">';
      echo '<a style="float:right;margin:5px" name="'.$this->iDebugNextLinkNumber.'Open" href="#'.$this->iDebugNextLinkNumber.'Close">Function Close '.$this->iDebugNextLinkNumber.'</a>';
      echo "<STRONG>{$fileName} : {$functionName}</STRONG>";
      echo '<hr style="clear:both">';
      array_push($this->aDebugOpenLinks, $this->iDebugNextLinkNumber);
      $this->iDebugNextLinkNumber++;
    }

    if ($this->bClassProfiling)
      $this->_ProfBegin($FunctionName);

    return TRUE;
  }
  
  /**
   * Called to end the debug run of a function.
   *
   * This method ends a <DIV><PRE> block and reports the time since $aStartTime
   * is clear to the debugging user.
   *
   * @author    Nigel Swinson <nigelswinson@users.sourceforge.net>
   * @param     $functionName (string) the name of the function we are beginning to debug
   * @param     $return_value (mixed) the return value from the function call that 
   *                                  we are debugging
   * @param     $bDebugFlag   (bool) TRUE if we are to draw a call stack, FALSE otherwise
   */
  function _closeDebugFunction($functionName, $returnValue = "", $bDebugFlag) {
    if ($bDebugFlag) {
      echo "<hr>";
      $iOpenLinkNumber = array_pop($this->aDebugOpenLinks);
      echo '<a style="float:right" name="'.$iOpenLinkNumber.'Close" href="#'.$iOpenLinkNumber.'Open">Function Open '.$iOpenLinkNumber.'</a>';
      if (isSet($returnValue)) {
      if (is_array($returnValue))
        echo "Return Value: ".print_r($returnValue)."\n";
      else if (is_numeric($returnValue)) 
        echo "Return Value: ".(string)$returnValue."\n";
      else if (is_bool($returnValue)) 
        echo "Return Value: ".($returnValue ? "TRUE" : "FALSE")."\n";
      else 
        echo "Return Value: \"".htmlspecialchars($returnValue)."\"\n";
    }
      echo '<br style="clear:both">';
      echo " \n</pre></div>";
    }
  
    if ($this->bClassProfiling)
      $this->_ProfEnd($FunctionName);

    return TRUE;
  }
  
  /**
    * Profile begin call
    */
  function _ProfBegin($sonFuncName) {
    static $entryTmpl = array ( 'start' => array(),
                  'recursiveCount' => 0,
                  'totTime' => 0,
                  'callCount' => 0  );
    $now = explode(' ', microtime());

    if (empty($this->callStack)) {
      $fatherFuncName = '';
    }
    else {
      $fatherFuncName = $this->callStack[sizeOf($this->callStack)-1];
      $fatherEntry = &$this->profile[$fatherFuncName];
    }
    $this->callStack[] = $sonFuncName;

    if (!isSet($this->profile[$sonFuncName])) {
      $this->profile[$sonFuncName] = $entryTmpl;
    }

    $sonEntry = &$this->profile[$sonFuncName];
    $sonEntry['callCount']++;
    // if we call the t's the same function let the time run, otherwise sum up
    if ($fatherFuncName == $sonFuncName) {
      $sonEntry['recursiveCount']++;
    }
    if (!empty($fatherFuncName)) {
      $last = $fatherEntry['start'];
    $fatherEntry['totTime'] += round( (($now[1] - $last[1]) + ($now[0] - $last[0]))*10000 );
      $fatherEntry['start'] = 0;
    }
    $sonEntry['start'] = explode(' ', microtime());
  }

  /**
   * Profile end call
   */
  function _ProfEnd($sonFuncName) {
    $now = explode(' ', microtime());

    array_pop($this->callStack);
    if (empty($this->callStack)) {
      $fatherFuncName = '';
    }
    else {
      $fatherFuncName = $this->callStack[sizeOf($this->callStack)-1];
      $fatherEntry = &$this->profile[$fatherFuncName];
    }
    $sonEntry = &$this->profile[$sonFuncName];
    if (empty($sonEntry)) {
      echo "ERROR in profEnd(): '$funcNam' not in list. Seams it was never started ;o)";
    }

    $last = $sonEntry['start'];
    $sonEntry['totTime'] += round( (($now[1] - $last[1]) + ($now[0] - $last[0]))*10000 );
    $sonEntry['start'] = 0;
    if (!empty($fatherEntry)) $fatherEntry['start'] = explode(' ', microtime());
  }

    /**
   * Show profile gathered so far as HTML table
   */
  function _ProfileToHtml() {
    $sortArr = array();
    if (empty($this->profile)) return '';
    reset($this->profile);
    while (list($funcName) = each($this->profile)) {
      $sortArrKey[] = $this->profile[$funcName]['totTime'];
      $sortArrVal[] = $funcName;
    }
    //echo '<pre>';var_dump($sortArrVal);echo '</pre>';
    array_multisort ($sortArrKey, SORT_DESC, $sortArrVal );
    //echo '<pre>';var_dump($sortArrVal);echo '</pre>';

    $totTime = 0;
    $size = sizeOf($sortArrVal);
    for ($i=0; $i<$size; $i++) {
      $funcName = &$sortArrVal[$i];
      $totTime += $this->profile[$funcName]['totTime'];
    }
    $out = '<table border="1">';
    $out .='<tr align="center" bgcolor="#bcd6f1"><th>Function</th><th> % </th><th>Total [ms]</th><th># Call</th><th>[ms] per Call</th><th># Recursive</th></tr>';
    for ($i=0; $i<$size; $i++) {
      $funcName = &$sortArrVal[$i];
      $row = &$this->profile[$funcName];
      $procent = round($row['totTime']*100/$totTime);
      if ($procent>20) $bgc = '#ff8080';
      elseif ($procent>15) $bgc = '#ff9999';
      elseif ($procent>10) $bgc = '#ffcccc';
      elseif ($procent>5) $bgc = '#ffffcc';
      else $bgc = '#66ff99';

      $out .="<tr align='center' bgcolor='{$bgc}'>";
      $out .='<td>'. $funcName .'</td><td>'. $procent .'% '.'</td><td>'. $row['totTime']/10 .'</td><td>'. $row['callCount'] .'</td><td>'. round($row['totTime']/10/$row['callCount'],2) .'</td><td>'. $row['recursiveCount'].'</td>';
      $out .='</tr>';
    }
    $out .= '</table> Total Time [' . $totTime/10 .'ms]' ;

    echo $out;
    return TRUE;
  }

  /**
   * Echo an XPath context for diagnostic purposes
   *
   * @param $context   (array)   An XPath context
   */
  function _printContext($context) {
    echo "{$context['nodePath']}({$context['pos']}/{$context['size']})";
  }
  
  /**
   * This is a debug helper function. It dumps the node-tree as HTML
   *
   * *QUICK AND DIRTY*. Needs some polishing.
   *
   * @param $node   (array)   A node 
   * @param $indent (string) (optional, default=''). For internal recursive calls.
   */
  function _treeDump($node, $indent = '') {
    $out = '';
    
    // Get rid of recursion
    $parentName = empty($node['parentNode']) ? "SUPER ROOT" :  $node['parentNode']['name'];
    unset($node['parentNode']);
    $node['parentNode'] = $parentName ;
    
    $out .= "NODE[{$node['name']}]\n";
    
    foreach($node as $key => $val) {
      if ($key === 'childNodes') continue;
      if (is_Array($val)) {
        $out .= $indent . "  [{$key}]\n" . arrayToStr($val, $indent . '    ');
      } else {
        $out .= $indent . "  [{$key}] => '{$val}' \n";
      }
    }
    
    if (!empty($node['childNodes'])) {
      $out .= $indent . "  ['childNodes'] (Size = ".sizeOf($node['childNodes']).")\n";
      foreach($node['childNodes'] as $key => $childNode) {
        $out .= $indent . "     [$key] => " . $this->_treeDump($childNode, $indent . '       ') . "\n";
      }
    }
    
    if (empty($indent)) {
      return "<pre>" . htmlspecialchars($out) . "</pre>";
    }
    return $out;
  }
} // END OF CLASS XPathBase


/************************************************************************************************
* ===============================================================================================
*                             X P a t h E n g i n e  -  Class                                    
* ===============================================================================================
************************************************************************************************/

class XPathEngine extends XPathBase {
  
  // List of supported XPath axes.
  // What a stupid idea from W3C to take axes name containing a '-' (dash)
  // NOTE: We replace the '-' with '_' to avoid the conflict with the minus operator.
  //       We will then do the same on the users Xpath querys
  //   -sibling => _sibling
  //   -or-     =>     _or_
  //  
  // This array contains a list of all valid axes that can be evaluated in an
  // XPath query.
  var $axes = array ( 'ancestor', 'ancestor_or_self', 'attribute', 'child', 'descendant', 
                        'descendant_or_self', 'following', 'following_sibling',  
                        'namespace', 'parent', 'preceding', 'preceding_sibling', 'self' 
     );
  
  // List of supported XPath functions.
  // What a stupid idea from W3C to take function name containing a '-' (dash)
  // NOTE: We replace the '-' with '_' to avoid the conflict with the minus operator.
  //       We will then do the same on the users Xpath querys 
  //   starts-with      => starts_with
  //   substring-before => substring_before
  //   substring-after  => substring_after
  //   string-length    => string_length
  //
  // This array contains a list of all valid functions that can be evaluated
  // in an XPath query.
  var $functions = array ( 'last', 'position', 'count', 'id', 'name',
    'string', 'concat', 'starts_with', 'contains', 'substring_before',
    'substring_after', 'substring', 'string_length', 'normalize_space', 'translate',
    'boolean', 'not', 'true', 'false', 'lang', 'number', 'sum', 'floor',
    'ceiling', 'round', 'x_lower', 'x_upper', 'generate_id' );
    
  // List of supported XPath operators.
  //
  // This array contains a list of all valid operators that can be evaluated
  // in a predicate of an XPath query. The list is ordered by the
  // precedence of the operators (lowest precedence first).
  var $operators = array( ' or ', ' and ', '=', '!=', '<=', '<', '>=', '>',
    '+', '-', '*', ' div ', ' mod ', ' | ');

  // List of literals from the xPath string.
  var $axPathLiterals = array();
  
  // The index and tree that is created during the analysis of an XML source.
  var $nodeIndex = array();
  var $nodeRoot  = array();
  var $emptyNode = array(
                     'name'        => '',       // The tag name. E.g. In <FOO bar="aaa"/> it would be 'FOO'
                     'attributes'  => array(),  // The attributes of the tag E.g. In <FOO bar="aaa"/> it would be array('bar'=>'aaa')
                     'childNodes'  => array(),  // Array of pointers to child nodes.
                     'textParts'   => array(),  // Array of text parts between the cilderen E.g. <FOO>aa<A>bb<B/>cc</A>dd</FOO> -> array('aa','bb','cc','dd')
                     'parentNode'   => NULL,     // Pointer to parent node or NULL if this node is the 'super root'
                     //-- *!* Following vars are set by the indexer and is for optimisation only *!*
                     'depth'       => 0,  // The tag depth (or tree level) starting with the root tag at 0.
                     'pos'         => 0,  // Is the zero-based position this node has in the parents 'childNodes'-list.
                     'contextPos'  => 1,  // Is the one-based position this node has by counting the siblings tags (tags with same name)
                     'xpath'       => ''  // Is the abs. XPath to this node.
                   );
  var $_indexIsDirty = FALSE;

  
  // These variable used during the parse XML source
  var $nodeStack       = array(); // The elements that we have still to close.
  var $parseStackIndex = 0;       // The current element of the nodeStack[] that we are adding to while 
                                  // parsing an XML source.  Corresponds to the depth of the xml node.
                                  // in our input data.
  var $parseOptions    = array(); // Used to set the PHP's XML parser options (see xml_parser_set_option)
  var $parsedTextLocation   = ''; // A reference to where we have to put char data collected during XML parsing
  var $parsInCData     = 0 ;      // Is >0 when we are inside a CDATA section.  
  var $parseSkipWhiteCache = 0;   // A cache of the skip whitespace parse option to speed up the parse.

  // This is the array of error strings, to keep consistency.
  var $errorStrings = array(
    'AbsoluteXPathRequired' => "The supplied xPath '%s' does not *uniquely* describe a node in the xml document.",
    'NoNodeMatch'           => "The supplied xPath-query '%s' does not match *any* node in the xml document.",
    'RootNodeAlreadyExists' => "An xml document may have only one root node."
    );
    
  /**
   * Constructor
   *
   * Optionally you may call this constructor with the XML-filename to parse and the 
   * XML option vector. Each of the entries in the option vector will be passed to
   * xml_parser_set_option().
   *
   * A option vector sample: 
   *   $xmlOpt = array(XML_OPTION_CASE_FOLDING => FALSE, 
   *                   XML_OPTION_SKIP_WHITE => TRUE);
   *
   * @param  $userXmlOptions (array) (optional) Vector of (<optionID>=><value>, 
   *                                 <optionID>=><value>, ...).  See PHP's
   *                                 xml_parser_set_option() docu for a list of possible
   *                                 options.
   * @see   importFromFile(), importFromString(), setXmlOptions()
   */
  function XPathEngine($userXmlOptions=array()) {
    parent::XPathBase();
    // Default to not folding case
    $this->parseOptions[XML_OPTION_CASE_FOLDING] = FALSE;
    // And not skipping whitespace
    $this->parseOptions[XML_OPTION_SKIP_WHITE] = FALSE;
    
    // Now merge in the overrides.
    // Don't use PHP's array_merge!
    if (is_array($userXmlOptions)) {
      foreach($userXmlOptions as $key => $val) $this->parseOptions[$key] = $val;
    }
  }
  
  /**
   * Resets the object so it's able to take a new xml sting/file
   *
   * Constructing objects is slow.  If you can, reuse ones that you have used already
   * by using this reset() function.
   */
  function reset() {
    parent::reset();
    $this->properties['xmlFile']  = ''; 
    $this->parseStackIndex = 0;
    $this->parsedTextLocation = '';
    $this->parsInCData   = 0;
    $this->nodeIndex     = array();
    $this->nodeRoot      = array();
    $this->nodeStack     = array();
    $this->aLiterals     = array();
    $this->_indexIsDirty = FALSE;
  }
  
  
  //-----------------------------------------------------------------------------------------
  // XPathEngine              ------  Get / Set Stuff  ------                                
  //-----------------------------------------------------------------------------------------
  
  /**
   * Returns the property/ies you want.
   * 
   * if $param is not given, all properties will be returned in a hash.
   *
   * @param  $param (string) the property you want the value of, or NULL for all the properties
   * @return        (mixed)  string OR hash of all params, or NULL on an unknown parameter.
   */
  function getProperties($param=NULL) {
    $this->properties['hasContent']      = !empty($this->nodeRoot);
    $this->properties['caseFolding']     = $this->parseOptions[XML_OPTION_CASE_FOLDING];
    $this->properties['skipWhiteSpaces'] = $this->parseOptions[XML_OPTION_SKIP_WHITE];
    
    if (empty($param)) return $this->properties;
    
    if (isSet($this->properties[$param])) {
      return $this->properties[$param];
    } else {
      return NULL;
    }
  }
  
  /**
   * Set an xml_parser_set_option()
   *
   * @param $optionID (int) The option ID (e.g. XML_OPTION_SKIP_WHITE)
   * @param $value    (int) The option value.
   * @see XML parser functions in PHP doc
   */
  function setXmlOption($optionID, $value) {
    if (!is_numeric($optionID)) return;
     $this->parseOptions[$optionID] = $value;
  }

  /**
   * Sets a number of xml_parser_set_option()s
   *
   * @param  $userXmlOptions (array) An array of parser options.
   * @see setXmlOption
   */
  function setXmlOptions($userXmlOptions=array()) {
    if (!is_array($userXmlOptions)) return;
    foreach($userXmlOptions as $key => $val) {
      $this->setXmlOption($key, $val);
    }
  }
  
  /**
   * Alternative way to control whether case-folding is enabled for this XML parser.
   *
   * Short cut to setXmlOptions(XML_OPTION_CASE_FOLDING, TRUE/FALSE)
   *
   * When it comes to XML, case-folding simply means uppercasing all tag- 
   * and attribute-names (NOT the content) if set to TRUE.  Note if you
   * have this option set, then your XPath queries will also be case folded 
   * for you.
   *
   * @param $onOff (bool) (default TRUE) 
   * @see XML parser functions in PHP doc
   */
  function setCaseFolding($onOff=TRUE) {
    $this->parseOptions[XML_OPTION_CASE_FOLDING] = $onOff;
  }
  
  /**
   * Alternative way to control whether skip-white-spaces is enabled for this XML parser.
   *
   * Short cut to setXmlOptions(XML_OPTION_SKIP_WHITE, TRUE/FALSE)
   *
   * When it comes to XML, skip-white-spaces will trim the tag content.
   * An XML file with no whitespace will be faster to process, but will make 
   * your data less human readable when you come to write it out.
   *
   * Running with this option on will slow the class down, so if you want to 
   * speed up your XML, then run it through once skipping white-spaces, then
   * write out the new version of your XML without whitespace, then use the
   * new XML file with skip whitespaces turned off.
   *
   * @param $onOff (bool) (default TRUE) 
   * @see XML parser functions in PHP doc
   */
  function setSkipWhiteSpaces($onOff=TRUE) {
    $this->parseOptions[XML_OPTION_SKIP_WHITE] = $onOff;
  }
   
  /**
   * Get the node defined by the $absoluteXPath.
   *
   * @param   $absoluteXPath (string) (optional, default is 'super-root') xpath to the node.
   * @return                 (array)  The node, or FALSE if the node wasn't found.
   */
  function &getNode($absoluteXPath='') {
    if ($absoluteXPath==='/') $absoluteXPath = '';
    if (!isSet($this->nodeIndex[$absoluteXPath])) return FALSE;
    if ($this->_indexIsDirty) $this->reindexNodeTree();
    return $this->nodeIndex[$absoluteXPath];
  }

  /**
   * Get a the content of a node text part or node attribute.
   * 
   * If the absolute Xpath references an attribute (Xpath ends with @ or attribute::), 
   * then the text value of that node-attribute is returned.
   * Otherwise the Xpath is referencing a text part of the node. This can be either a 
   * direct reference to a text part (Xpath ends with text()[<nr>]) or indirect reference 
   * (a simple abs. Xpath to a node).
   * 1) Direct Reference (xpath ends with text()[<part-number>]):
   *   If the 'part-number' is omitted, the first text-part is assumed; starting by 1.
   *   Negative numbers are allowed, where -1 is the last text-part a.s.o.
   * 2) Indirect Reference (a simple abs. Xpath to a node):
   *   Default is to return the *whole text*; that is the concated text-parts of the matching
   *   node. (NOTE that only in this case you'll only get a copy and changes to the returned  
   *   value wounld have no effect). Optionally you may pass a parameter 
   *   $textPartNr to define the text-part you want;  starting by 1.
   *   Negative numbers are allowed, where -1 is the last text-part a.s.o.
   *
   * NOTE I : The returned value can be fetched by reference
   *          E.g. $text =& wholeText(). If you wish to modify the text.
   * NOTE II: text-part numbers out of range will return FALSE
   * SIDENOTE:The function name is a suggestion from W3C in the XPath specification level 3.
   *
   * @param   $absoluteXPath  (string)  xpath to the node (See above).
   * @param   $textPartNr     (int)     If referring to a node, specifies which text part 
   *                                    to query.
   * @return                  (&string) A *reference* to the text if the node that the other 
   *                                    parameters describe or FALSE if the node is not found.
   */
  function &wholeText($absoluteXPath, $textPartNr=NULL) {
    $status = FALSE;
    $text   = NULL;
    if ($this->_indexIsDirty) $this->reindexNodeTree();
    
    do { // try-block
      if (preg_match(";(.*)/(attribute::|@)([^/]*)$;U", $absoluteXPath, $matches)) {
        $absoluteXPath = $matches[1];
        $attribute = $matches[3];
        if (!isSet($this->nodeIndex[$absoluteXPath]['attributes'][$attribute])) {
          $this->_displayError("The $absoluteXPath/attribute::$attribute value isn't a node in this document.", __LINE__, __FILE__, FALSE);
          break; // try-block
        }
        $text =& $this->nodeIndex[$absoluteXPath]['attributes'][$attribute];
        $status = TRUE;
        break; // try-block
      }
            
      // Xpath contains a 'text()'-function, thus goes right to a text node. If so interpret the Xpath.
      if (preg_match(":(.*)/text\(\)(\[(.*)\])?$:U", $absoluteXPath, $matches)) {
        $absoluteXPath = $matches[1];
 
        if (!isSet($this->nodeIndex[$absoluteXPath])) {
            $this->_displayError("The $absoluteXPath value isn't a node in this document.", __LINE__, __FILE__, FALSE);
            break; // try-block
        }

        // Get the amount of the text parts in the node.
        $textPartSize = sizeOf($this->nodeIndex[$absoluteXPath]['textParts']);

        // default to the first text node if a text node was not specified
        $textPartNr = isSet($matches[2]) ? substr($matches[2],1,-1) : 1;

        // Support negative indexes like -1 === last a.s.o.
        if ($textPartNr < 0) $textPartNr = $textPartSize + $textPartNr +1;
        if (($textPartNr <= 0) OR ($textPartNr > $textPartSize)) {
          $this->_displayError("The $absoluteXPath/text()[$textPartNr] value isn't a NODE in this document.", __LINE__, __FILE__, FALSE);
          break; // try-block
        }
        $text =& $this->nodeIndex[$absoluteXPath]['textParts'][$textPartNr - 1];
        $status = TRUE;
        break; // try-block
      }
      
      // At this point we have been given an xpath with neither a 'text()' nor 'attribute::' axis at the end
      // So we assume a get to text is wanted and use the optioanl fallback parameters $textPartNr
     
      if (!isSet($this->nodeIndex[$absoluteXPath])) {
          $this->_displayError("The $absoluteXPath value isn't a node in this document.", __LINE__, __FILE__, FALSE);
          break; // try-block
      }

      // Get the amount of the text parts in the node.
      $textPartSize…

Large files files are truncated, but you can click here to view the full file