highwire /
opensource-php-BetterDOMDocument
Duplicate code is one of the most pungent code smells. A rule that is often used is to re-structure code once it is duplicated in three or more places.
Common duplication problems, and corresponding solutions are:
Complex classes like DOMDoc often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use DOMDoc, and based on these observations, apply Extract Interface, too.
| 1 | <?php namespace BetterDOMDocument; |
||
| 11 | class DOMDoc extends \DOMDocument { |
||
| 12 | |||
| 13 | private $auto_ns = FALSE; |
||
| 14 | public $ns = array(); |
||
| 15 | public $default_ns = FALSE; |
||
| 16 | public $error_checking = 'strict'; // Can be 'strict', 'warning', 'none' / FALSE |
||
| 17 | |||
| 18 | /** |
||
| 19 | * Create a new DOMDoc |
||
| 20 | * |
||
| 21 | * @param mixed $xml |
||
| 22 | * $xml can either be an XML string, a DOMDocument, or a DOMElement. |
||
| 23 | * You can also pass FALSE or NULL (or omit it) and load XML later using loadXML or loadHTML |
||
| 24 | * |
||
| 25 | * @param mixed $auto_register_namespaces |
||
| 26 | * Auto-register namespaces. All namespaces in the root element will be registered for use in xpath queries. |
||
| 27 | * Namespaces that are not declared in the root element will not be auto-registered |
||
| 28 | * Defaults to TRUE (Meaning it will auto register all auxiliary namespaces but not the default namespace). |
||
| 29 | * Pass a prefix string to automatically register the default namespace. |
||
| 30 | * Pass FALSE to disable auto-namespace registeration |
||
| 31 | * |
||
| 32 | * @param bool $error_checking |
||
| 33 | * Can be 'strict', 'warning', or 'none. Defaults to 'strict'. |
||
| 34 | * 'none' supresses all errors |
||
| 35 | * 'warning' is the default behavior in DOMDocument |
||
| 36 | * 'strict' corresponds to DOMDocument strictErrorChecking TRUE |
||
| 37 | */ |
||
| 38 | 24 | public function __construct($xml = FALSE, $auto_register_namespaces = TRUE, $error_checking = 'strict') { |
|
| 73 | |||
| 74 | /** |
||
| 75 | * Register a namespace to be used in xpath queries |
||
| 76 | * |
||
| 77 | * @param string $prefix |
||
| 78 | * Namespace prefix to register |
||
| 79 | * |
||
| 80 | * @param string $url |
||
| 81 | * Connonical URL for this namespace prefix |
||
| 82 | */ |
||
| 83 | 13 | public function registerNamespace($prefix, $url) { |
|
| 86 | |||
| 87 | /** |
||
| 88 | * Get the list of registered namespaces as an array |
||
| 89 | */ |
||
| 90 | 7 | public function getNamespaces() { |
|
| 93 | |||
| 94 | /** |
||
| 95 | * Given a namespace URL, get the prefix |
||
| 96 | * |
||
| 97 | * @param string $url |
||
| 98 | * Connonical URL for this namespace prefix |
||
| 99 | * |
||
| 100 | * @return string|false |
||
| 101 | * The namespace prefix or FALSE if there is no namespace with that URL |
||
| 102 | */ |
||
| 103 | 1 | public function lookupPrefix($url) { |
|
| 104 | 1 | return array_search($url, $this->ns); |
|
| 105 | } |
||
| 106 | |||
| 107 | /** |
||
| 108 | * Given a namespace prefix, get the URL |
||
| 109 | * |
||
| 110 | * @param string $prefix |
||
| 111 | * namespace prefix |
||
| 112 | * |
||
| 113 | * return string|false |
||
| 114 | * The namespace URL or FALSE if there is no namespace with that prefix |
||
| 115 | */ |
||
| 116 | 1 | public function lookupURL($prefix) { |
|
| 117 | 1 | if (isset($this->ns[$prefix])) { |
|
| 118 | 1 | return $this->ns[$prefix]; |
|
| 119 | } |
||
| 120 | else { |
||
| 121 | return FALSE; |
||
| 122 | } |
||
| 123 | } |
||
| 124 | |||
| 125 | /** |
||
| 126 | * Given an xpath, get a list of nodes. |
||
| 127 | * |
||
| 128 | * @param string $xpath |
||
| 129 | * xpath to be used for query |
||
| 130 | * |
||
| 131 | * @param mixed $context |
||
| 132 | * $context can either be an xpath string, or a DOMElement |
||
| 133 | * Provides context for the xpath query |
||
| 134 | * |
||
| 135 | * @return DOMList|false |
||
| 136 | * A DOMList object, which is very similar to a DOMNodeList, but with better iterabilility. |
||
| 137 | */ |
||
| 138 | 18 | public function xpath($xpath, $context = NULL) { |
|
| 166 | |||
| 167 | |||
| 168 | /** |
||
| 169 | * Given an xpath, get a single node (first one found) |
||
| 170 | * |
||
| 171 | * @param string $xpath |
||
| 172 | * xpath to be used for query |
||
| 173 | * |
||
| 174 | * @param mixed $context |
||
| 175 | * $context can either be an xpath string, or a DOMElement |
||
| 176 | * Provides context for the xpath query |
||
| 177 | * |
||
| 178 | * @return mixed |
||
| 179 | * The first node found by the xpath query |
||
| 180 | */ |
||
| 181 | 18 | public function xpathSingle($xpath, $context = NULL) { |
|
| 191 | |||
| 192 | |||
| 193 | /** |
||
| 194 | * Given an CSS selector, get a list of nodes. |
||
| 195 | * |
||
| 196 | * @param string $css_selector |
||
| 197 | * CSS Selector to be used for query |
||
| 198 | * |
||
| 199 | * @param mixed $context |
||
| 200 | * $context can either be an xpath string, or a DOMElement |
||
| 201 | * Provides context for the CSS selector |
||
| 202 | * |
||
| 203 | * @return DOMList|false |
||
| 204 | * A DOMList object, which is very similar to a DOMNodeList, but with better iterabilility. |
||
| 205 | */ |
||
| 206 | 1 | public function select($css_selector, $context = NULL) { |
|
| 207 | 1 | $converter = new CssSelectorConverter(); |
|
| 208 | 1 | $xpath = $converter->toXPath($css_selector); |
|
| 209 | |||
| 210 | 1 | return $this->xpath($xpath, $context); |
|
| 211 | } |
||
| 212 | |||
| 213 | /** |
||
| 214 | * Given an CSS selector, get a single node. |
||
| 215 | * |
||
| 216 | * @param string $css_selector |
||
| 217 | * CSS Selector to be used for query |
||
| 218 | * |
||
| 219 | * @param mixed $context |
||
| 220 | * $context can either be an xpath string, or a DOMElement |
||
| 221 | * Provides context for the CSS selector |
||
| 222 | * |
||
| 223 | * @return DOMList |
||
| 224 | * A DOMList object, which is very similar to a DOMNodeList, but with better iterabilility. |
||
| 225 | */ |
||
| 226 | 1 | public function selectSingle($css_selector, $context = NULL) { |
|
| 227 | 1 | $converter = new CssSelectorConverter(); |
|
| 228 | 1 | $xpath = $converter->toXPath($css_selector); |
|
| 229 | |||
| 230 | 1 | return $this->xpathSingle($xpath, $context); |
|
| 231 | } |
||
| 232 | |||
| 233 | /** |
||
| 234 | * Get the document (or an element) as an array |
||
| 235 | * |
||
| 236 | * @param string $raw |
||
| 237 | * Can be either FALSE, 'full', or 'inner'. Defaults to FALSE. |
||
| 238 | * When set to 'full' every node's full XML is also attached to the array |
||
| 239 | * When set to 'inner' every node's inner XML is attached to the array. |
||
| 240 | * |
||
| 241 | * @param mixed $context |
||
| 242 | * Optional context node. Can pass an DOMElement object or an xpath string. |
||
| 243 | * If passed, only the given node will be used when generating the array |
||
| 244 | */ |
||
| 245 | 1 | public function getArray($raw = FALSE, $context = NULL) { |
|
| 246 | 1 | $array = false; |
|
| 247 | |||
| 248 | 1 | $this->createContext($context, 'xpath', FALSE); |
|
| 249 | |||
| 250 | 1 | if ($context) { |
|
| 251 | 1 | if ($raw == 'full') { |
|
| 252 | 1 | $array['#raw'] = $this->saveXML($context); |
|
| 253 | } |
||
| 254 | 1 | if ($raw == 'inner') { |
|
| 255 | 1 | $array['#raw'] = $this->innerText($context); |
|
| 256 | } |
||
| 257 | 1 | if ($context->hasAttributes()) { |
|
| 258 | 1 | foreach ($context->attributes as $attr) { |
|
| 259 | 1 | $array['@'.$attr->nodeName] = $attr->nodeValue; |
|
| 260 | } |
||
| 261 | } |
||
| 262 | |||
| 263 | 1 | if ($context->hasChildNodes()) { |
|
| 264 | 1 | if ($context->childNodes->length == 1 && $context->firstChild->nodeType == XML_TEXT_NODE) { |
|
| 265 | 1 | $array['#text'] = $context->firstChild->nodeValue; |
|
| 266 | } |
||
| 267 | else { |
||
| 268 | 1 | foreach ($context->childNodes as $childNode) { |
|
| 269 | 1 | if ($childNode->nodeType == XML_ELEMENT_NODE) { |
|
| 270 | 1 | $array[$childNode->nodeName][] = $this->getArray($raw, $childNode); |
|
| 271 | } |
||
| 272 | 1 | elseif ($childNode->nodeType == XML_CDATA_SECTION_NODE) { |
|
| 273 | 1 | $array['#text'] = $childNode->textContent; |
|
| 274 | } |
||
| 275 | } |
||
| 276 | } |
||
| 277 | } |
||
| 278 | } |
||
| 279 | // Else no node was passed, which means we are processing the entire domDocument |
||
| 280 | else { |
||
| 281 | 1 | foreach ($this->childNodes as $childNode) { |
|
| 282 | 1 | if ($childNode->nodeType == XML_ELEMENT_NODE) { |
|
| 283 | 1 | $array[$childNode->nodeName][] = $this->getArray($raw, $childNode); |
|
| 284 | } |
||
| 285 | } |
||
| 286 | } |
||
| 287 | |||
| 288 | 1 | return $array; |
|
| 289 | } |
||
| 290 | |||
| 291 | /** |
||
| 292 | * Get the inner text of an element |
||
| 293 | * |
||
| 294 | * @param mixed $context |
||
| 295 | * Optional context node. Can pass an DOMElement object or an xpath string. |
||
| 296 | */ |
||
| 297 | 1 | public function innerText($context = NULL) { |
|
| 298 | 1 | $this->createContext($context, 'xpath'); |
|
| 299 | |||
| 300 | 1 | $pattern = "/<".preg_quote($context->nodeName)."\b[^>]*>(.*)<\/".preg_quote($context->nodeName).">/s"; |
|
| 301 | 1 | $matches = array(); |
|
| 302 | 1 | if (preg_match($pattern, $this->saveXML($context), $matches)) { |
|
| 303 | 1 | return $matches[1]; |
|
| 304 | } |
||
| 305 | else { |
||
| 306 | 1 | return ''; |
|
| 307 | } |
||
| 308 | } |
||
| 309 | |||
| 310 | /** |
||
| 311 | * Create an DOMElement from XML and attach it to the DOMDocument |
||
| 312 | * |
||
| 313 | * Note that this does not place it anywhere in the dom tree, it merely imports it. |
||
| 314 | * |
||
| 315 | * @param string $xml |
||
| 316 | * XML string to import |
||
| 317 | */ |
||
| 318 | 5 | public function createElementFromXML($xml) { |
|
| 350 | |||
| 351 | /** |
||
| 352 | * Append a child to the context node, make it the last child |
||
| 353 | * |
||
| 354 | * @param mixed $newnode |
||
| 355 | * $newnode can either be an XML string, a DOMDocument, or a DOMElement. |
||
| 356 | * |
||
| 357 | * @param mixed $context |
||
| 358 | * $context can either be an xpath string, or a DOMElement |
||
| 359 | * Omiting $context results in using the root document element as the context |
||
| 360 | * |
||
| 361 | * @return DOMElement|false |
||
| 362 | * The $newnode, properly attached to DOMDocument. If you passed $newnode as a DOMElement |
||
| 363 | * then you should replace your DOMElement with the returned one. |
||
| 364 | */ |
||
| 365 | 1 | public function append($newnode, $context = NULL) { |
|
| 382 | |||
| 383 | /** |
||
| 384 | * Append a child to the context node, make it the first child |
||
| 385 | * |
||
| 386 | * @param mixed $newnode |
||
| 387 | * $newnode can either be an XML string, a DOMDocument, or a DOMElement. |
||
| 388 | * |
||
| 389 | * @param mixed $context |
||
| 390 | * $context can either be an xpath string, or a DOMElement |
||
| 391 | * Omiting $context results in using the root document element as the context |
||
| 392 | * |
||
| 393 | * @return DOMElement|false |
||
| 394 | * The $newnode, properly attached to DOMDocument. If you passed $newnode as a DOMElement |
||
| 395 | * then you should replace your DOMElement with the returned one. |
||
| 396 | */ |
||
| 397 | 1 | View Code Duplication | public function prepend($newnode, $context = NULL) { |
| 407 | |||
| 408 | /** |
||
| 409 | * Prepend a sibling to the context node, put it just before the context node |
||
| 410 | * |
||
| 411 | * @param mixed $newnode |
||
| 412 | * $newnode can either be an XML string, a DOMDocument, or a DOMElement. |
||
| 413 | * |
||
| 414 | * @param mixed $context |
||
| 415 | * $context can either be an xpath string, or a DOMElement |
||
| 416 | * Omiting $context results in using the root document element as the context |
||
| 417 | * |
||
| 418 | * @return DOMElement|false |
||
| 419 | * The $newnode, properly attached to DOMDocument. If you passed $newnode as a DOMElement |
||
| 420 | * then you should replace your DOMElement with the returned one. |
||
| 421 | */ |
||
| 422 | 1 | View Code Duplication | public function prependSibling($newnode, $context = NULL) { |
| 432 | |||
| 433 | /** |
||
| 434 | * Append a sibling to the context node, put it just after the context node |
||
| 435 | * |
||
| 436 | * @param mixed $newnode |
||
| 437 | * $newnode can either be an XML string, a DOMDocument, or a DOMElement. |
||
| 438 | * |
||
| 439 | * @param mixed $context |
||
| 440 | * $context can either be an xpath string, or a DOMElement |
||
| 441 | * Omiting $context results in using the root document element as the context |
||
| 442 | * |
||
| 443 | * @return DOMElement|false |
||
| 444 | * The $newnode, properly attached to DOMDocument. If you passed $newnode as a DOMElement |
||
| 445 | * then you should replace your DOMElement with the returned one. |
||
| 446 | */ |
||
| 447 | 1 | public function appendSibling($newnode, $context) { |
|
| 470 | |||
| 471 | /** |
||
| 472 | * Given an xpath or DOMElement, return a new DOMDoc. |
||
| 473 | * |
||
| 474 | * @param mixed $node |
||
| 475 | * $node can either be an xpath string or a DOMElement. |
||
| 476 | * |
||
| 477 | * @return DOMDoc |
||
| 478 | * A new DOMDoc created from the xpath or DOMElement |
||
| 479 | */ |
||
| 480 | 7 | public function extract($node, $auto_register_namespaces = TRUE, $error_checking = 'none') { |
|
| 486 | |||
| 487 | /** |
||
| 488 | * Given a pair of nodes, replace the first with the second |
||
| 489 | * |
||
| 490 | * @param mixed $node |
||
| 491 | * Node to be replaced. Can either be an xpath string or a DOMDocument (or even a DOMNode). |
||
| 492 | * |
||
| 493 | * @param mixed $replace |
||
| 494 | * Replace $node with $replace. Replace can be an XML string, or a DOMNode |
||
| 495 | * |
||
| 496 | * @return mixed |
||
| 497 | * The overwritten / replaced node. |
||
| 498 | */ |
||
| 499 | 2 | public function replace($node, $replace) { |
|
| 500 | 2 | $this->createContext($node, 'xpath'); |
|
| 501 | 2 | $this->createContext($replace, 'xml'); |
|
| 502 | |||
| 503 | 2 | if (!$node || !$replace) { |
|
| 504 | return FALSE; |
||
| 505 | } |
||
| 506 | |||
| 507 | 2 | if (!$replace->ownerDocument->documentElement->isSameNode($this->documentElement)) { |
|
| 508 | 1 | $replace = $this->importNode($replace, true); |
|
| 509 | } |
||
| 510 | 2 | $node->parentNode->replaceChild($replace, $node); |
|
| 511 | 2 | $node = $replace; |
|
| 512 | 2 | return $node; |
|
| 513 | } |
||
| 514 | |||
| 515 | /** |
||
| 516 | * Given a node(s), remove / delete them |
||
| 517 | * |
||
| 518 | * @param mixed $node |
||
| 519 | * Can pass a DOMNode, a NodeList, DOMNodeList, an xpath string, or an array of any of these. |
||
| 520 | */ |
||
| 521 | 1 | public function remove($node) { |
|
| 542 | |||
| 543 | /** |
||
| 544 | * Given an XSL string, transform the DOMDoc (or a passed context node) |
||
| 545 | * |
||
| 546 | * @param string $xsl |
||
| 547 | * XSL Transormation |
||
| 548 | * |
||
| 549 | * @param mixed $context |
||
| 550 | * $context can either be an xpath string, or a DOMElement. Ommiting it |
||
| 551 | * results in transforming the entire document |
||
| 552 | * |
||
| 553 | * @return a new DOMDoc |
||
| 554 | */ |
||
| 555 | 4 | public function tranform($xsl, $context = NULL) { |
|
| 572 | |||
| 573 | /** |
||
| 574 | * Given a node, change it's namespace to the specified namespace in situ |
||
| 575 | * |
||
| 576 | * @param mixed $node |
||
| 577 | * Node to be changed. Can either be an xpath string or a DOMElement. |
||
| 578 | * |
||
| 579 | * @param mixed $prefix |
||
| 580 | * prefix for the new namespace |
||
| 581 | * |
||
| 582 | * @param mixed $url |
||
| 583 | * The URL for the new namespace |
||
| 584 | * |
||
| 585 | * @return mixed |
||
| 586 | * The node with the new namespace. The node will also be changed in-situ in the document as well. |
||
| 587 | */ |
||
| 588 | 1 | public function changeNamespace($node, $prefix, $url) { |
|
| 589 | 1 | $this->createContext($node, 'xpath'); |
|
| 590 | |||
| 591 | 1 | if (!$node) { |
|
| 592 | return FALSE; |
||
| 593 | } |
||
| 594 | |||
| 595 | 1 | $this->registerNamespace($prefix, $url); |
|
| 596 | |||
| 597 | 1 | if (get_class($node) == 'DOMElement') { |
|
| 598 | $xsl = ' |
||
| 599 | <xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"> |
||
| 600 | <xsl:template match="*"> |
||
| 601 | 1 | <xsl:element name="' . $prefix . ':{local-name()}" namespace="' . $url . '"> |
|
| 602 | <xsl:copy-of select="@*"/> |
||
| 603 | <xsl:apply-templates/> |
||
| 604 | </xsl:element> |
||
| 605 | </xsl:template> |
||
| 606 | 1 | </xsl:stylesheet>'; |
|
| 607 | |||
| 608 | 1 | $transformed = $this->tranform($xsl, $node); |
|
| 609 | 1 | return $this->replace($node, $transformed->documentElement); |
|
| 610 | } |
||
| 611 | else { |
||
| 612 | // @@TODO: Report the correct calling file and number |
||
| 613 | throw new Exception("Changing the namespace of a " . get_class($node) . " is not supported"); |
||
| 614 | } |
||
| 615 | } |
||
| 616 | |||
| 617 | /** |
||
| 618 | * Get a lossless HTML representation of the XML |
||
| 619 | * |
||
| 620 | * Transforms the document (or passed context) into a set of HTML spans. |
||
| 621 | * The element name becomes the class, all other attributes become HTML5 |
||
| 622 | * "data-" attributes. |
||
| 623 | * |
||
| 624 | * @param mixed $context |
||
| 625 | * $context can either be an xpath string, or a DOMElement. Ommiting it |
||
| 626 | * results in transforming the entire document |
||
| 627 | * |
||
| 628 | * @param array $options |
||
| 629 | * Options for transforming the HTML into XML. The following options are supported: |
||
| 630 | * 'xlink' => {TRUE or xpath} |
||
| 631 | * Transform xlink links into <a href> elements. If you specify 'xlink' => TRUE then |
||
| 632 | * it will transform all elements with xlink:type = simple into a <a href> element. |
||
| 633 | * Alternatively you may specify your own xpath for selecting which elements get transformed |
||
| 634 | * into <a href> tags. |
||
| 635 | * @return HTML string |
||
| 636 | */ |
||
| 637 | 3 | public function asHTML($context = NULL, $options = array()) { |
|
| 709 | |||
| 710 | /** |
||
| 711 | * Output the DOMDoc as an XML string |
||
| 712 | * |
||
| 713 | * @param mixed $context |
||
| 714 | * $context can either be an xpath string, or a DOMElement. Ommiting it |
||
| 715 | * results in outputting the entire document |
||
| 716 | * |
||
| 717 | * @return XML string |
||
| 718 | */ |
||
| 719 | 12 | public function out($context = NULL) { |
|
| 744 | |||
| 745 | /** |
||
| 746 | * Magic method for casting a DOMDoc as a string |
||
| 747 | */ |
||
| 748 | 1 | public function __toString() { |
|
| 751 | |||
| 752 | 24 | public function setErrorChecking($error_checking) { |
|
| 764 | |||
| 765 | 14 | public static function loadFile($file_or_url, $auto_register_namespaces = TRUE) { |
|
| 773 | |||
| 774 | 1 | public function loadHTML($source, $options = NULL) { |
|
| 775 | 1 | $success = parent::loadHTML($source, $options); |
|
| 776 | 1 | $this->AutoRegisterNamespace(TRUE); |
|
| 777 | |||
| 778 | 1 | return boolval($success); |
|
| 779 | } |
||
| 780 | |||
| 781 | 14 | public function loadXML($source, $options = NULL) { |
|
| 787 | |||
| 788 | 24 | private function AutoRegisterNamespace($auto_register_namespaces) { |
|
| 838 | |||
| 839 | 22 | private function createContext(&$context, $type = 'xpath', $createDocument = TRUE) { |
|
| 869 | } |
||
| 870 | |||
| 871 | |||
| 873 |
