lib/xmlize.php

   1 <?php
   2 // This file is part of Moodle - http://moodle.org/
   3 //
   4 // Moodle is free software: you can redistribute it and/or modify
   5 // it under the terms of the GNU General Public License as published by
   6 // the Free Software Foundation, either version 3 of the License, or
   7 // (at your option) any later version.
   8 //
   9 // Moodle is distributed in the hope that it will be useful,
  10 // but WITHOUT ANY WARRANTY; without even the implied warranty of
  11 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12 // GNU General Public License for more details.
  13 //
  14 // You should have received a copy of the GNU General Public License
  15 // along with Moodle.  If not, see <http://www.gnu.org/licenses/>.
  16
  17 /**
  18  * Code for parsing xml files.
  19  *
  20  * Handles functionality for:
  21  *
  22  * Import of xml files in questionbank and course import.
  23  * Can handle xml files larger than 10MB through chunking the input file.
  24  * Replaces the original xmlize by Hans Anderson, {@link http://www.hansanderson.com/contact/}
  25  * with equal interface.
  26  *
  27  * @package    core
  28  * @subpackage lib
  29  * @copyright  Kilian Singer
  30  * @license    http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
  31  */
  32
  33 /**
  34  * Exception thrown when there is an error parsing an XML file.
  35  *
  36  * @copyright 2010 The Open University
  37  * @license    http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
  38  */
  39
  40 defined('MOODLE_INTERNAL') || die();
  41
  42 class xml_format_exception extends moodle_exception {
  43     /** @var string */
  44     public $errorstring;
  45     /** @var string */
  46     public $char;
  47     /**
  48      * Constructor function
  49      *
  50      * @param string $errorstring Errorstring
  51      * @param int $line Linenumber
  52      * @param string $char Errorcharacter
  53      * @param string $link Link
  54      */
  55     public function __construct($errorstring, $line, $char, $link = '') {
  56         $this->errorstring = $errorstring;
  57         $this->line = $line;
  58         $this->char = $char;
  59
  60         $a = new stdClass();
  61         $a->errorstring = $errorstring;
  62         $a->errorline = $line;
  63         $a->errorchar = $char;
  64         parent::__construct('errorparsingxml', 'error', $link, $a);
  65     }
  66 }
  67
  68 /**
  69  * Class for parsing xml files.
  70  *
  71  * Handles functionality for:
  72  *
  73  * Import of xml files in questionbank and course import.
  74  * Can handle xml files larger than 10MB through chunking the input file.
  75  * Uses a similar interface to the original version xmlize() by Hans Anderson.
  76  *
  77  * @package    core
  78  * @subpackage lib
  79  * @copyright  Kilian Singer
  80  * @license    http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
  81  */
  82 class core_xml_parser {
  83     /** @var array resulting $xml tree */
  84     private $xml;
  85     /** @var array stores references to constructed $xml tree */
  86     private $current;
  87     /** @var int tores the level in the XML tree */
  88     private $level;
  89     /**
  90      * Is called when tags are opened.
  91      *
  92      * Note: Used by xml element handler as callback.
  93      *
  94      * @author Kilian Singer
  95      * @param resource $parser The XML parser resource.
  96      * @param string $name The XML source to parse.
  97      * @param array $attrs Stores attributes of XML tag.
  98      */
  99     private function startelement($parser, $name, $attrs) {
 100         $current = &$this->current;
 101         $level = &$this->level;
 102         if (!empty($name)) {
 103             if ($level == 0) {
 104                 $current[$level][$name] = array();
 105                 $current[$level][$name]["@"] = $attrs; // Attribute.
 106                 $current[$level][$name]["#"] = array(); // Other tags.
 107                 $current[$level + 1] = & $current[$level][$name]["#"];
 108                 $level++;
 109             } else {
 110                 if (empty($current[$level][$name])) {
 111                     $current[$level][$name] = array();
 112                 }
 113                 $siz = count($current[$level][$name]);
 114                 if (!empty($attrs)) {
 115                     $current[$level][$name][$siz]["@"] = $attrs; // Attribute.
 116                 }
 117                 $current[$level][$name][$siz]["#"] = array(); // Other tags.
 118                 $current[$level + 1] = & $current[$level][$name][$siz]["#"];
 119                 $level++;
 120             }
 121         }
 122     }
 123
 124     /**
 125      * Is called when tags are closed.
 126      *
 127      * Note: Used by xml element handler as callback.
 128      *
 129      * @author Kilian Singer
 130      * @param resource $parser The XML parser resource.
 131      * @param string $name The XML source to parse.
 132      */
 133     private function endelement($parser, $name) {
 134         $current = &$this->current;
 135         $level = &$this->level;
 136         if (!empty($name)) {
 137             if (empty($current[$level])) {
 138                 $current[$level] = '';
 139             } else if (array_key_exists(0, $current[$level])) {
 140                 if (count($current[$level]) == 1) {
 141                     $current[$level] = $current[$level][0]; // We remove array index if we only have a single entry.
 142                 }
 143             }
 144
 145             $level--;
 146         }
 147     }
 148     /**
 149      * Is called for text between the start and the end of tags.
 150      *
 151      * Note: Used by xml element handler as callback.
 152      *
 153      * @author Kilian Singer
 154      * @param resource $parser The XML parser resource.
 155      * @param string $data The XML source to parse.
 156      */
 157     private function characterdata($parser, $data) {
 158         $current = &$this->current;
 159         $level = &$this->level;
 160         if (($data == "0") || (!empty($data) && trim($data) != "")) {
 161             $siz = count($current[$level]);
 162             if ($siz == 0) {
 163                 $current[$level][0] = $data;
 164             } else {
 165                 $key = max(array_keys($current[$level]));
 166                 if (is_int($key)) {
 167                     end($current[$level]);
 168                     if (is_int(key($current[$level]))) { // If last index is nummeric we have CDATA and concat.
 169                         $current[$level][$key] = $current[$level][$key] . $data;
 170                     } else {
 171                         $current[$level][$key + 1] = $data; // Otherwise we make a new key.
 172                     }
 173                 } else {
 174                     $current[$level][0] = $data;
 175                 }
 176             }
 177         }
 178     }
 179
 180     /**
 181      * Parses XML string.
 182      *
 183      * Note: Interface is kept equal to previous version.
 184      *
 185      * @author Kilian Singer
 186      * @param string $data the XML source to parse.
 187      * @param int $whitespace If set to 1 allows the parser to skip "space" characters in xml document. Default is 1
 188      * @param string $encoding Specify an OUTPUT encoding. If not specified, it defaults to UTF-8.
 189      * @param bool $reporterrors if set to true, then a {@link xml_format_exception}
 190      *      exception will be thrown if the XML is not well-formed. Otherwise errors are ignored.
 191      * @return array|false representation of the parsed XML.
 192      */
 193     public function parse($data, $whitespace = 1, $encoding = 'UTF-8', $reporterrors = false) {
 194         $data = trim($data);
 195         $this->xml = array();
 196         $this->current = array();
 197         $this->level = 0;
 198         $this->current[0] = & $this->xml;
 199         $parser = xml_parser_create($encoding);
 200         xml_parser_set_option($parser, XML_OPTION_CASE_FOLDING, 0);
 201         xml_parser_set_option($parser, XML_OPTION_SKIP_WHITE, $whitespace);
 202         xml_set_element_handler($parser, [$this, "startelement"], [$this, "endelement"]);
 203         xml_set_character_data_handler($parser, [$this, "characterdata"]);
 204         // Start parsing an xml document.
 205         for ($i = 0; $i < strlen($data); $i += 4096) {
 206             if (!xml_parse($parser, substr($data, $i, 4096))) {
 207                 break;
 208             }
 209         }
 210         if ($reporterrors) {
 211             $errorcode = xml_get_error_code($parser);
 212             if ($errorcode) {
 213                 $exception = new xml_format_exception(xml_error_string($errorcode),
 214                         xml_get_current_line_number($parser),
 215                         xml_get_current_column_number($parser));
 216                 xml_parser_free($parser);
 217                 throw $exception;
 218             }
 219         }
 220         xml_parser_free($parser); // Deletes the parser.
 221         if (empty($this->xml)) { // XML file is invalid or empty, return false.
 222             return false;
 223         }
 224         return $this->xml;
 225     }
 226 }
 227
 228 /**
 229  * XML parsing function calles into class.
 230  *
 231  * Note: Used by xml element handler as callback.
 232  *
 233  * @param string $data the XML source to parse.
 234  * @param int $whitespace If set to 1 allows the parser to skip "space" characters in xml document. Default is 1
 235  * @param string $encoding Specify an OUTPUT encoding. If not specified, it defaults to UTF-8.
 236  * @param bool $reporterrors if set to true, then a {@link xml_format_exception}
 237  *      exception will be thrown if the XML is not well-formed. Otherwise errors are ignored.
 238  * @return array representation of the parsed XML.
 239  */
 240 function xmlize($data, $whitespace = 1, $encoding = 'UTF-8', $reporterrors = false) {
 241     $hxml = new core_xml_parser();
 242     return $hxml->parse($data, $whitespace, $encoding, $reporterrors);
 243 }