1: <?php
2: /**
3: * XML handling for Cake.
4: *
5: * The methods in these classes enable the datasources that use XML to work.
6: *
7: * CakePHP(tm) : Rapid Development Framework (http://cakephp.org)
8: * Copyright (c) Cake Software Foundation, Inc. (http://cakefoundation.org)
9: *
10: * Licensed under The MIT License
11: * For full copyright and license information, please see the LICENSE.txt
12: * Redistributions of files must retain the above copyright notice.
13: *
14: * @copyright Copyright (c) Cake Software Foundation, Inc. (http://cakefoundation.org)
15: * @link http://cakephp.org CakePHP(tm) Project
16: * @package Cake.Utility
17: * @since CakePHP v .0.10.3.1400
18: * @license http://www.opensource.org/licenses/mit-license.php MIT License
19: */
20:
21: App::uses('HttpSocket', 'Network/Http');
22:
23: /**
24: * XML handling for CakePHP.
25: *
26: * The methods in these classes enable the datasources that use XML to work.
27: *
28: * @package Cake.Utility
29: */
30: class Xml {
31:
32: /**
33: * Initialize SimpleXMLElement or DOMDocument from a given XML string, file path, URL or array.
34: *
35: * ### Usage:
36: *
37: * Building XML from a string:
38: *
39: * `$xml = Xml::build('<example>text</example>');`
40: *
41: * Building XML from string (output DOMDocument):
42: *
43: * `$xml = Xml::build('<example>text</example>', array('return' => 'domdocument'));`
44: *
45: * Building XML from a file path:
46: *
47: * `$xml = Xml::build('/path/to/an/xml/file.xml');`
48: *
49: * Building from a remote URL:
50: *
51: * `$xml = Xml::build('http://example.com/example.xml');`
52: *
53: * Building from an array:
54: *
55: * ```
56: * $value = array(
57: * 'tags' => array(
58: * 'tag' => array(
59: * array(
60: * 'id' => '1',
61: * 'name' => 'defect'
62: * ),
63: * array(
64: * 'id' => '2',
65: * 'name' => 'enhancement'
66: * )
67: * )
68: * )
69: * );
70: * $xml = Xml::build($value);
71: * ```
72: *
73: * When building XML from an array ensure that there is only one top level element.
74: *
75: * ### Options
76: *
77: * - `return` Can be 'simplexml' to return object of SimpleXMLElement or 'domdocument' to return DOMDocument.
78: * - `loadEntities` Defaults to false. Set to true to enable loading of `<!ENTITY` definitions. This
79: * is disabled by default for security reasons.
80: * - `readFile` Set to false to disable file reading. This is important to disable when
81: * putting user data into Xml::build(). If enabled local & remote files will be read if they exist.
82: * Defaults to true for backwards compatibility reasons.
83: * - If using array as input, you can pass `options` from Xml::fromArray.
84: *
85: * @param string|array $input XML string, a path to a file, a URL or an array
86: * @param array $options The options to use
87: * @return SimpleXMLElement|DOMDocument SimpleXMLElement or DOMDocument
88: * @throws XmlException
89: */
90: public static function build($input, $options = array()) {
91: if (!is_array($options)) {
92: $options = array('return' => (string)$options);
93: }
94: $defaults = array(
95: 'return' => 'simplexml',
96: 'loadEntities' => false,
97: 'readFile' => true
98: );
99: $options += $defaults;
100:
101: if (is_array($input) || is_object($input)) {
102: return self::fromArray((array)$input, $options);
103: } elseif (strpos($input, '<') !== false) {
104: return self::_loadXml($input, $options);
105: } elseif ($options['readFile'] && file_exists($input)) {
106: return self::_loadXml(file_get_contents($input), $options);
107: } elseif ($options['readFile'] && strpos($input, 'http://') === 0 || strpos($input, 'https://') === 0) {
108: try {
109: $socket = new HttpSocket(array('request' => array('redirect' => 10)));
110: $response = $socket->get($input);
111: if (!$response->isOk()) {
112: throw new XmlException(__d('cake_dev', 'XML cannot be read.'));
113: }
114: return self::_loadXml($response->body, $options);
115: } catch (SocketException $e) {
116: throw new XmlException(__d('cake_dev', 'XML cannot be read.'));
117: }
118: } elseif (!is_string($input)) {
119: throw new XmlException(__d('cake_dev', 'Invalid input.'));
120: }
121: throw new XmlException(__d('cake_dev', 'XML cannot be read.'));
122: }
123:
124: /**
125: * Parse the input data and create either a SimpleXmlElement object or a DOMDocument.
126: *
127: * @param string $input The input to load.
128: * @param array $options The options to use. See Xml::build()
129: * @return SimpleXmlElement|DOMDocument
130: * @throws XmlException
131: */
132: protected static function _loadXml($input, $options) {
133: $hasDisable = function_exists('libxml_disable_entity_loader');
134: $internalErrors = libxml_use_internal_errors(true);
135: if ($hasDisable && !$options['loadEntities']) {
136: libxml_disable_entity_loader(true);
137: }
138: try {
139: if ($options['return'] === 'simplexml' || $options['return'] === 'simplexmlelement') {
140: $xml = new SimpleXMLElement($input, LIBXML_NOCDATA);
141: } else {
142: $xml = new DOMDocument();
143: $xml->loadXML($input);
144: }
145: } catch (Exception $e) {
146: $xml = null;
147: }
148: if ($hasDisable && !$options['loadEntities']) {
149: libxml_disable_entity_loader(false);
150: }
151: libxml_use_internal_errors($internalErrors);
152: if ($xml === null) {
153: throw new XmlException(__d('cake_dev', 'Xml cannot be read.'));
154: }
155: return $xml;
156: }
157:
158: /**
159: * Transform an array into a SimpleXMLElement
160: *
161: * ### Options
162: *
163: * - `format` If create childs ('tags') or attributes ('attributes').
164: * - `pretty` Returns formatted Xml when set to `true`. Defaults to `false`
165: * - `version` Version of XML document. Default is 1.0.
166: * - `encoding` Encoding of XML document. If null remove from XML header. Default is the some of application.
167: * - `return` If return object of SimpleXMLElement ('simplexml') or DOMDocument ('domdocument'). Default is SimpleXMLElement.
168: *
169: * Using the following data:
170: *
171: * ```
172: * $value = array(
173: * 'root' => array(
174: * 'tag' => array(
175: * 'id' => 1,
176: * 'value' => 'defect',
177: * '@' => 'description'
178: * )
179: * )
180: * );
181: * ```
182: *
183: * Calling `Xml::fromArray($value, 'tags');` Will generate:
184: *
185: * `<root><tag><id>1</id><value>defect</value>description</tag></root>`
186: *
187: * And calling `Xml::fromArray($value, 'attributes');` Will generate:
188: *
189: * `<root><tag id="1" value="defect">description</tag></root>`
190: *
191: * @param array $input Array with data
192: * @param array $options The options to use
193: * @return SimpleXMLElement|DOMDocument SimpleXMLElement or DOMDocument
194: * @throws XmlException
195: */
196: public static function fromArray($input, $options = array()) {
197: if (!is_array($input) || count($input) !== 1) {
198: throw new XmlException(__d('cake_dev', 'Invalid input.'));
199: }
200: $key = key($input);
201: if (is_int($key)) {
202: throw new XmlException(__d('cake_dev', 'The key of input must be alphanumeric'));
203: }
204:
205: if (!is_array($options)) {
206: $options = array('format' => (string)$options);
207: }
208: $defaults = array(
209: 'format' => 'tags',
210: 'version' => '1.0',
211: 'encoding' => Configure::read('App.encoding'),
212: 'return' => 'simplexml',
213: 'pretty' => false
214: );
215: $options += $defaults;
216:
217: $dom = new DOMDocument($options['version'], $options['encoding']);
218: if ($options['pretty']) {
219: $dom->formatOutput = true;
220: }
221: self::_fromArray($dom, $dom, $input, $options['format']);
222:
223: $options['return'] = strtolower($options['return']);
224: if ($options['return'] === 'simplexml' || $options['return'] === 'simplexmlelement') {
225: return new SimpleXMLElement($dom->saveXML());
226: }
227: return $dom;
228: }
229:
230: /**
231: * Recursive method to create childs from array
232: *
233: * @param DOMDocument $dom Handler to DOMDocument
234: * @param DOMElement $node Handler to DOMElement (child)
235: * @param array &$data Array of data to append to the $node.
236: * @param string $format Either 'attributes' or 'tags'. This determines where nested keys go.
237: * @return void
238: * @throws XmlException
239: */
240: protected static function _fromArray($dom, $node, &$data, $format) {
241: if (empty($data) || !is_array($data)) {
242: return;
243: }
244: foreach ($data as $key => $value) {
245: if (is_string($key)) {
246: if (!is_array($value)) {
247: if (is_bool($value)) {
248: $value = (int)$value;
249: } elseif ($value === null) {
250: $value = '';
251: }
252: $isNamespace = strpos($key, 'xmlns:');
253: if ($isNamespace !== false) {
254: $node->setAttributeNS('http://www.w3.org/2000/xmlns/', $key, $value);
255: continue;
256: }
257: if ($key[0] !== '@' && $format === 'tags') {
258: $child = null;
259: if (!is_numeric($value)) {
260: // Escape special characters
261: // http://www.w3.org/TR/REC-xml/#syntax
262: // https://bugs.php.net/bug.php?id=36795
263: $child = $dom->createElement($key, '');
264: $child->appendChild(new DOMText($value));
265: } else {
266: $child = $dom->createElement($key, $value);
267: }
268: $node->appendChild($child);
269: } else {
270: if ($key[0] === '@') {
271: $key = substr($key, 1);
272: }
273: $attribute = $dom->createAttribute($key);
274: $attribute->appendChild($dom->createTextNode($value));
275: $node->appendChild($attribute);
276: }
277: } else {
278: if ($key[0] === '@') {
279: throw new XmlException(__d('cake_dev', 'Invalid array'));
280: }
281: if (is_numeric(implode('', array_keys($value)))) { // List
282: foreach ($value as $item) {
283: $itemData = compact('dom', 'node', 'key', 'format');
284: $itemData['value'] = $item;
285: self::_createChild($itemData);
286: }
287: } else { // Struct
288: self::_createChild(compact('dom', 'node', 'key', 'value', 'format'));
289: }
290: }
291: } else {
292: throw new XmlException(__d('cake_dev', 'Invalid array'));
293: }
294: }
295: }
296:
297: /**
298: * Helper to _fromArray(). It will create childs of arrays
299: *
300: * @param array $data Array with informations to create childs
301: * @return void
302: */
303: protected static function _createChild($data) {
304: extract($data);
305: $childNS = $childValue = null;
306: if (is_array($value)) {
307: if (isset($value['@'])) {
308: $childValue = (string)$value['@'];
309: unset($value['@']);
310: }
311: if (isset($value['xmlns:'])) {
312: $childNS = $value['xmlns:'];
313: unset($value['xmlns:']);
314: }
315: } elseif (!empty($value) || $value === 0) {
316: $childValue = (string)$value;
317: }
318:
319: $child = $dom->createElement($key);
320: if ($childValue !== null) {
321: $child->appendChild($dom->createTextNode($childValue));
322: }
323: if ($childNS) {
324: $child->setAttribute('xmlns', $childNS);
325: }
326:
327: self::_fromArray($dom, $child, $value, $format);
328: $node->appendChild($child);
329: }
330:
331: /**
332: * Returns this XML structure as an array.
333: *
334: * @param SimpleXMLElement|DOMDocument|DOMNode $obj SimpleXMLElement, DOMDocument or DOMNode instance
335: * @return array Array representation of the XML structure.
336: * @throws XmlException
337: */
338: public static function toArray($obj) {
339: if ($obj instanceof DOMNode) {
340: $obj = simplexml_import_dom($obj);
341: }
342: if (!($obj instanceof SimpleXMLElement)) {
343: throw new XmlException(__d('cake_dev', 'The input is not instance of SimpleXMLElement, DOMDocument or DOMNode.'));
344: }
345: $result = array();
346: $namespaces = array_merge(array('' => ''), $obj->getNamespaces(true));
347: self::_toArray($obj, $result, '', array_keys($namespaces));
348: return $result;
349: }
350:
351: /**
352: * Recursive method to toArray
353: *
354: * @param SimpleXMLElement $xml SimpleXMLElement object
355: * @param array &$parentData Parent array with data
356: * @param string $ns Namespace of current child
357: * @param array $namespaces List of namespaces in XML
358: * @return void
359: */
360: protected static function _toArray($xml, &$parentData, $ns, $namespaces) {
361: $data = array();
362:
363: foreach ($namespaces as $namespace) {
364: foreach ($xml->attributes($namespace, true) as $key => $value) {
365: if (!empty($namespace)) {
366: $key = $namespace . ':' . $key;
367: }
368: $data['@' . $key] = (string)$value;
369: }
370:
371: foreach ($xml->children($namespace, true) as $child) {
372: self::_toArray($child, $data, $namespace, $namespaces);
373: }
374: }
375:
376: $asString = trim((string)$xml);
377: if (empty($data)) {
378: $data = $asString;
379: } elseif (strlen($asString) > 0) {
380: $data['@'] = $asString;
381: }
382:
383: if (!empty($ns)) {
384: $ns .= ':';
385: }
386: $name = $ns . $xml->getName();
387: if (isset($parentData[$name])) {
388: if (!is_array($parentData[$name]) || !isset($parentData[$name][0])) {
389: $parentData[$name] = array($parentData[$name]);
390: }
391: $parentData[$name][] = $data;
392: } else {
393: $parentData[$name] = $data;
394: }
395: }
396:
397: }
398: