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 int */
  46     public $line;
  47     /** @var char */
  48     public $char;
  49     /**
  50      * Constructor function
  51      *
  52      * @param string $errorstring Errorstring
  53      * @param int $line Linenumber
  54      * @param char $char Errorcharacter
  55      * @param string $link Link
  56      */
  57     public function __construct($errorstring, $line, $char, $link = '') {
  58         $this->errorstring = $errorstring;
  59         $this->line = $line;
  60         $this->char = $char;
  61
  62         $a = new stdClass();
  63         $a->errorstring = $errorstring;
  64         $a->errorline = $line;
  65         $a->errorchar = $char;
  66         parent::__construct('errorparsingxml', 'error', $link, $a);
  67     }
  68 }
  69
  70 /**
  71  * Class for parsing xml files.
  72  *
  73  * Handles functionality for:
  74  *
  75  * Import of xml files in questionbank and course import.
  76  * Can handle xml files larger than 10MB through chunking the input file.
  77  * Uses a similar interface to the original version xmlize() by Hans Anderson.
  78  *
  79  * @package    core
  80  * @subpackage lib
  81  * @copyright  Kilian Singer
  82  * @license    http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
  83  */
  84 class core_xml_parser {
  85     /** @var array resulting $xml tree */
  86     private $xml;
  87     /** @var array stores references to constructed $xml tree */
  88     private $current;
  89     /** @var int tores the level in the XML tree */
  90     private $level;
  91     /**
  92      * Is called when tags are opened.
  93      *
  94      * Note: Used by xml element handler as callback.
  95      *
  96      * @author Kilian Singer
  97      * @param resource $parser The XML parser resource.
  98      * @param string $name The XML source to parse.
  99      * @param array $attrs Stores attributes of XML tag.
 100      */
 101     private function startelement($parser, $name, $attrs) {
 102         $current = &$this->current;
 103         $level = &$this->level;
 104         if (!empty($name)) {
 105             if ($level == 0) {
 106                 $current[$level][$name] = array();
 107                 $current[$level][$name]["@"] = $attrs; // Attribute.
 108                 $current[$level][$name]["#"] = array(); // Other tags.
 109                 $current[$level + 1] = & $current[$level][$name]["#"];
 110                 $level++;
 111             } else {
 112                 if (empty($current[$level][$name])) {
 113                     $current[$level][$name] = array();
 114                 }
 115                 $siz = count($current[$level][$name]);
 116                 if (!empty($attrs)) {
 117                     $current[$level][$name][$siz]["@"] = $attrs; // Attribute.
 118                 }
 119                 $current[$level][$name][$siz]["#"] = array(); // Other tags.
 120                 $current[$level + 1] = & $current[$level][$name][$siz]["#"];
 121                 $level++;
 122             }
 123         }
 124     }
 125
 126     /**
 127      * Is called when tags are closed.
 128      *
 129      * Note: Used by xml element handler as callback.
 130      *
 131      * @author Kilian Singer
 132      * @param resource $parser The XML parser resource.
 133      * @param string $name The XML source to parse.
 134      */
 135     private function endelement($parser, $name) {
 136         $current = &$this->current;
 137         $level = &$this->level;
 138         if (!empty($name)) {
 139             if (empty($current[$level])) {
 140                 $current[$level] = '';
 141             } else if (array_key_exists(0, $current[$level])) {
 142                 if (count($current[$level]) == 1) {
 143                     $current[$level] = $current[$level][0]; // We remove array index if we only have a single entry.
 144                 }
 145             }
 146
 147             $level--;
 148         }
 149     }
 150     /**
 151      * Is called for text between the start and the end of tags.
 152      *
 153      * Note: Used by xml element handler as callback.
 154      *
 155      * @author Kilian Singer
 156      * @param resource $parser The XML parser resource.
 157      * @param string $data The XML source to parse.
 158      */
 159     private function characterdata($parser, $data) {
 160         $current = &$this->current;
 161         $level = &$this->level;
 162         if (($data == "0") || (!empty($data) && trim($data) != "")) {
 163             $siz = count($current[$level]);
 164             if ($siz == 0) {
 165                 $current[$level][0] = $data;
 166             } else {
 167                 $key = max(array_keys($current[$level]));
 168                 if (is_int($key)) {
 169                     end($current[$level]);
 170                     if (is_int(key($current[$level]))) { // If last index is nummeric we have CDATA and concat.
 171                         $current[$level][$key] = $current[$level][$key] . $data;
 172                     } else {
 173                         $current[$level][$key + 1] = $data; // Otherwise we make a new key.
 174                     }
 175                 } else {
 176                     $current[$level][0] = $data;
 177                 }
 178             }
 179         }
 180     }
 181
 182     /**
 183      * Parses XML string.
 184      *
 185      * Note: Interface is kept equal to previous version.
 186      *
 187      * @author Kilian Singer
 188      * @param string $data the XML source to parse.
 189      * @param int $whitespace If set to 1 allows the parser to skip "space" characters in xml document. Default is 1
 190      * @param string $encoding Specify an OUTPUT encoding. If not specified, it defaults to UTF-8.
 191      * @param bool $reporterrors if set to true, then a {@link xml_format_exception}
 192      *      exception will be thrown if the XML is not well-formed. Otherwise errors are ignored.
 193      * @return array representation of the parsed XML.
 194      */
 195     public function parse($data, $whitespace = 1, $encoding = 'UTF-8', $reporterrors = false) {
 196         $data = trim($data);
 197         $this->xml = array();
 198         $this->current = array();
 199         $this->level = 0;
 200         $this->current[0] = & $this->xml;
 201         $parser = xml_parser_create($encoding);
 202         xml_parser_set_option($parser, XML_OPTION_CASE_FOLDING, 0);
 203         xml_parser_set_option($parser, XML_OPTION_SKIP_WHITE, $whitespace);
 204         xml_set_element_handler($parser, [$this, "startelement"], [$this, "endelement"]);
 205         xml_set_character_data_handler($parser, [$this, "characterdata"]);
 206         // Start parsing an xml document.
 207         for ($i = 0; $i < strlen($data); $i += 4096) {
 208             if (!xml_parse($parser, substr($data, $i, 4096))) {
 209                 break;
 210             }
 211         }
 212         if ($reporterrors) {
 213             $errorcode = xml_get_error_code($parser);
 214             if ($errorcode) {
 215                 $exception = new xml_format_exception(xml_error_string($errorcode),
 216                         xml_get_current_line_number($parser),
 217                         xml_get_current_column_number($parser));
 218                 xml_parser_free($parser);
 219                 throw $exception;
 220             }
 221         }
 222         xml_parser_free($parser); // Deletes the parser.
 223         if (empty($this->xml)) { // XML file is invalid or empty, return false.
 224             return false;
 225         }
 226         return $this->xml;
 227     }
 228 }
 229
 230 /**
 231  * XML parsing function calles into class.
 232  *
 233  * Note: Used by xml element handler as callback.
 234  *
 235  * @param string $data the XML source to parse.
 236  * @param int $whitespace If set to 1 allows the parser to skip "space" characters in xml document. Default is 1
 237  * @param string $encoding Specify an OUTPUT encoding. If not specified, it defaults to UTF-8.
 238  * @param bool $reporterrors if set to true, then a {@link xml_format_exception}
 239  *      exception will be thrown if the XML is not well-formed. Otherwise errors are ignored.
 240  * @return array representation of the parsed XML.
 241  */
 242 function xmlize($data, $whitespace = 1, $encoding = 'UTF-8', $reporterrors = false) {
 243     $hxml = new core_xml_parser();
 244     return $hxml->parse($data, $whitespace, $encoding, $reporterrors);
 245 }