PHP DOMDocument::removeChild方法代码示例

 /**
  * Wrap a CAS 2.0 XML response and output it as a string.
  *
  * @return string CAS 2.0+ server response as an XML string.
  */
 public function prepare()
 {
     $root = $this->createElement('serviceResponse');
     if (!empty($this->response)) {
         $root->appendChild($this->response);
     }
     // Removing all child nodes from response document:
     while ($this->document->firstChild) {
         $this->document->removeChild($this->document->firstChild);
     }
     $this->document->appendChild($root);
     return $this->document->saveXML();
 }

/**
 * Tries to convert the given HTML into a plain text format - best suited for
 * e-mail display, etc.
 *
 * <p>In particular, it tries to maintain the following features:
 * <ul>
 *   <li>Links are maintained, with the 'href' copied over
 *   <li>Information in the &lt;head&gt; is lost
 * </ul>
 *
 * @param html the input HTML
 * @return the HTML converted, as best as possible, to text
 */
function convert_html_to_text($html, $width = 74)
{
    $html = fix_newlines($html);
    $doc = new DOMDocument('1.0', 'utf-8');
    if (strpos($html, '<?xml ') === false) {
        $html = '<?xml encoding="utf-8"?>' . $html;
    }
    # <?php (4vim)
    if (!@$doc->loadHTML($html)) {
        return $html;
    }
    // Thanks, http://us3.php.net/manual/en/domdocument.loadhtml.php#95251
    // dirty fix -- remove the inserted processing instruction
    foreach ($doc->childNodes as $item) {
        if ($item->nodeType == XML_PI_NODE) {
            $doc->removeChild($item);
            // remove hack
            break;
        }
    }
    $elements = identify_node($doc);
    // Add the default stylesheet
    $elements->getRoot()->addStylesheet(HtmlStylesheet::fromArray(array('html' => array('white-space' => 'pre'), 'p' => array('margin-bottom' => '1em'), 'pre' => array('white-space' => 'pre'))));
    $options = array();
    if (is_object($elements)) {
        $output = $elements->render($width, $options);
    } else {
        $output = $elements;
    }
    return trim($output);
}

 /**
  * {@inheritdoc}
  */
 public function parseHtml($html, $encoding = 'UTF-8')
 {
     $document = new \DOMDocument();
     foreach ($this->config as $name => $value) {
         $document->{$name} = $value;
     }
     $document->encoding = $encoding;
     if ($encoding !== false) {
         // Tell the parser which charset to use
         $encoding = $encoding ?: $document->encoding;
         $encoding = '<?xml encoding="' . $encoding . '" ?>';
         $html = $encoding . $html;
         // @codingStandardsIgnoreStart
         @$document->loadHTML($html);
         // @codingStandardsIgnoreEnd
         foreach ($document->childNodes as $item) {
             if ($item->nodeType == XML_PI_NODE) {
                 $document->removeChild($item);
             }
         }
     } else {
         // @codingStandardsIgnoreStart
         @$document->loadHTML($html);
         // @codingStandardsIgnoreEnd
     }
     return $document;
 }

 static function fixChildrenClass($content, $elementType, $class, $fixRawHtml = true)
 {
     $classPath = '\\AbstractElement\\' . $elementType;
     // is this an object that extends AbstractElement?
     if (is_a($content, '\\Element')) {
         // is this of the right element type?
         if (is_a($content, $classPath)) {
             $content->addClass($class);
         }
         if (isset($content->contents)) {
             foreach ($content->contents as $index => $value) {
                 $content->contents[$index] = AbstractElement\Helper::fixChildrenClass($value, $elementType, $class);
             }
         }
         return $content;
     }
     if (is_string($content) && $fixRawHtml) {
         $dom = new \DOMDocument();
         $dom->loadHtml($content);
         $reflectionClass = new \ReflectionClass($classPath);
         $elements = $dom->getElementsByTagName($reflectionClass->getConstant('tag'));
         foreach ($elements as $element) {
             $element->setAttribute('class', $element->getAttribute('class') . ' ' . $class);
         }
         $dom->removeChild($dom->doctype);
         $dom->replaceChild($dom->firstChild->firstChild, $dom->firstChild);
         return $dom->saveHTML();
     }
 }

 function parse($content)
 {
     $doc = new DOMDocument();
     @$doc->loadHTML("<div>" . $content . "</div>");
     $links = $doc->getElementsByTagName('a');
     $blogurl = get_bloginfo("url");
     $components = parse_url($blogurl);
     $host = $components["host"];
     /* Set target attribute of all external links to "_blank" */
     foreach ($links as $link) {
         $href = $link->getAttribute("href");
         $components = parse_url($href);
         if (!isset($components["host"])) {
             continue;
         }
         if ($components["host"] != $host) {
             $link->setAttribute("target", "_blank");
         }
     }
     /*
      * Extract the HTML fragment.
      * Credits: http://stackoverflow.com/questions/29493678/loadhtml-libxml-html-noimplied-on-an-html-fragment-generates-incorrect-tags
      */
     $temporary_wrapper = $doc->getElementsByTagName('div')->item(0);
     $temporary_wrapper = $temporary_wrapper->parentNode->removeChild($temporary_wrapper);
     while ($doc->firstChild) {
         $doc->removeChild($doc->firstChild);
     }
     while ($temporary_wrapper->firstChild) {
         $doc->appendChild($temporary_wrapper->firstChild);
     }
     /* Return HTML */
     return trim($doc->saveHTML());
 }

 function preg_replace_html($content, $tags)
 {
     $dom = new DOMDocument();
     if (!@$dom->loadHTML('<?xml encoding="UTF-8">' . $content)) {
         return $content;
     }
     foreach ($dom->childNodes as $item) {
         if ($item->nodeType === XML_PI_NODE) {
             $dom->removeChild($item);
             break;
         }
     }
     $dom->encoding = 'UTF-8';
     $images = $dom->getElementsByTagName('img');
     $blankImage = 'data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7';
     for ($i = $images->length - 1; $i >= 0; $i--) {
         $node = $images->item($i);
         $clone = $node->cloneNode();
         $noscript = $dom->createElement('noscript');
         $noscript->appendChild($clone);
         $node->parentNode->insertBefore($noscript, $node);
         $node->setAttribute('data-src', $node->getAttribute('src'));
         $node->setAttribute('src', $blankImage);
         $node->setAttribute('class', trim($node->getAttribute('class') . ' lazy'));
     }
     $newHtml = $dom->saveHTML();
     if (!$newHtml) {
         return $content;
     }
     return $newHtml;
 }

 /**
  * Convert HTML into a textual representation
  *
  * @return string Text representation
  */
 public function convert()
 {
     if (!empty($this->text)) {
         return $this->text;
     }
     $html = $this->preprocess($this->html);
     $this->document = new DOMDocument();
     set_error_handler(array(__CLASS__, 'silence_errors'));
     $this->document->loadHTML($html);
     restore_error_handler();
     // Remove the DOCTYPE
     // Seems to cause segfaulting if we don't do this
     if ($this->document->firstChild instanceof DOMDocumentType) {
         $this->document->removeChild($this->document->firstChild);
     }
     $this->text = $this->parse_children($this->document->getElementsByTagName('body')->item(0));
     $this->text = preg_replace("#\n{3,}#", "\n\n", $this->text);
     $this->text = $this->wrap($this->text, 80);
     $this->text .= $this->generate_links();
     return $this->text;
 }

 public function __construct($document, $tagname = '*')
 {
     if (!is_scalar($document)) {
         throw new Exception('Not a valid {JQMDoc} object');
     }
     $this->_namespace = microtime(true) . uniqid();
     jqm_var($this->_namespace, $this);
     //$document = preg_replace('/\s+/',' ',$document);
     //Detect if $document is a valid full document
     $hasHTML = stripos($document, '<html') !== false;
     $this->__documentMap = array();
     $DOM = new DOMDocument();
     $DOM->recover = true;
     $DOM->preserveWhiteSpace = true;
     $DOM->substituteEntities = true;
     $DOM->formatOutput = true;
     $DOM->encoding = 'utf-8';
     $DOM->loadHTML(mb_convert_encoding($document, 'HTML-ENTITIES', 'UTF-8'));
     $DOM->normalizeDocument();
     $html = $DOM->getElementsByTagName($tagname);
     //Determine root / pieced map
     $hasRoot = false;
     if ($html->item(0)->childNodes->length > 0) {
         $hasRoot = false;
         $html_tmp = $html;
         if ($html->item(0)->tagName == 'html') {
             $html_tmp = $html->item(0)->childNodes->item(0);
         }
         if (!$hasHTML and $html_tmp->childNodes->length == 1) {
             $hasRoot = true;
             if ($html_tmp->childNodes->item(0)->firstChild) {
                 $root = $html_tmp->childNodes->item(0)->firstChild->getNodePath();
             } else {
                 $root = $html_tmp->childNodes->item(0)->getNodePath();
             }
         }
     }
     $this->__schema['root'] = $hasRoot;
     $this->__schema['rootPath'] = $root;
     $this->__mapLength = false;
     $this->_length = false;
     $this->length = false;
     if ($DOM->doctype) {
         //$this->__schema['doctype'] = $DOM->saveHTML($DOM->doctype);
         $DOM->removeChild($DOM->doctype);
     }
     //$output = $DOM->saveHTML();
     //$this->__documentRaw = $output;
     $this->_DOM = $DOM;
     $this->_selector = $tagname;
 }

    function hook_article_filter($article)
    {
        if (defined('NO_CURL') || !function_exists("curl_init")) {
            return $article;
        }
        $charset_hack = '<head>
			<meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
		</head>';
        $doc = new DOMDocument();
        $doc->loadHTML($charset_hack . $article["content"]);
        $found = false;
        if ($doc) {
            $xpath = new DOMXpath($doc);
            $images = $xpath->query('(//img[@src])');
            foreach ($images as $img) {
                $src = $img->getAttribute("src");
                $ch = curl_init($src);
                curl_setopt($ch, CURLOPT_HEADER, 0);
                curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
                curl_setopt($ch, CURLOPT_BINARYTRANSFER, 1);
                curl_setopt($ch, CURLOPT_RANGE, "0-32768");
                @($result = curl_exec($ch));
                $http_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
                if ($result && ($http_code == 200 || $http_code == 206)) {
                    $filename = tempnam(sys_get_temp_dir(), "ttsizecheck");
                    if ($filename) {
                        $fh = fopen($filename, "w");
                        if ($fh) {
                            fwrite($fh, $result);
                            fclose($fh);
                            @($info = getimagesize($filename));
                            if ($info && $info[0] > 0 && $info[1] > 0) {
                                $img->setAttribute("width", $info[0]);
                                $img->setAttribute("height", $info[1]);
                                $found = true;
                            }
                            unlink($filename);
                        }
                    }
                }
            }
            if ($found) {
                $doc->removeChild($doc->firstChild);
                //remove doctype
                $article["content"] = $doc->saveHTML();
            }
        }
        return $article;
    }

 /**
  * @param string $item
  * @param int $key
  * @return string
  */
 public function extProc_beforeAllWrap($item, $key)
 {
     if (!empty($item)) {
         $pageId = $this->I['uid'];
         $dom = new \DOMDocument();
         $dom->loadHTML(mb_convert_encoding($item, 'HTML-ENTITIES', 'UTF-8'));
         $link = $dom->getElementsByTagName('a');
         $item = $link->item(0);
         $dataAttribute = 'bwrk_onepage_' . $pageId;
         $classAttribute = $dom->createAttribute('data-bwrkonepage-id');
         $classAttribute->value = $dataAttribute;
         $item->appendChild($classAttribute);
         $dom->removeChild($dom->doctype);
         $dom->replaceChild($dom->firstChild->firstChild->firstChild, $dom->firstChild);
         $newItem = $dom->saveHTML();
         return $newItem;
     }
 }

 /**
  * Append product recommendations to Autocomplete block html
  * 
  * @param string $html
  * @return string
  */
 protected function _appendTopRecommendations($html)
 {
     $recommendationsModel = Mage::getModel('autocompleterecommendations/recommendation');
     $query = Mage::helper('catalogsearch')->getQuery();
     $productRecommendationsHtml = $recommendationsModel->getProductRecommendationsHtml($query);
     if ($productRecommendationsHtml) {
         $dom = new DOMDocument('1.0', 'utf8');
         $dom->loadHTML($html);
         $uls = $dom->getElementsByTagName('ul');
         $ul = $uls->item(0);
         $productRecommendationsDom = $recommendationsModel->getRecommendationsDom($productRecommendationsHtml);
         $productRecommendationsDom = $dom->importNode($productRecommendationsDom, true);
         $ul->appendChild($productRecommendationsDom);
         $dom->removeChild($dom->doctype);
         $dom->replaceChild($dom->firstChild->firstChild->firstChild, $dom->firstChild);
         $html = $dom->saveHTML();
     }
     return $html;
 }

function fof_item_targets($content)
{
    /* quiet warnings */
    $old_xml_err = libxml_use_internal_errors(true);
    $dom = new DOMDocument();
    /*
    	Load content into DOM, within a div wrapper.  Wrapper div will be
    	stripped before returning altered content.  Without doing this,
    	any bare text content would get wrapped in p elements while being
    	parsed in.
    */
    $dom->loadHtml('<div>' . mb_convert_encoding($content, 'HTML-ENTITIES', "UTF-8") . '</div>');
    /* strip <!DOCTYPE> which DOMDocument adds */
    $dom->removeChild($dom->firstChild);
    /* strip <html><body> which DOMDocument adds */
    $dom->replaceChild($dom->firstChild->firstChild->firstChild, $dom->firstChild);
    /* replace or add link targets */
    $xpath = new DOMXpath($dom);
    foreach ($xpath->query('//a') as $node) {
        $node->setAttribute('target', '_blank');
    }
    $content_out = '';
    /* emit the updated contents inside our div */
    /* start at the first node inside first div.. */
    $node = $dom->firstChild->firstChild;
    while ($node) {
        $content_out .= $dom->saveHTML($node);
        /* repeat for all nodes at this level */
        $node = $node->nextSibling;
    }
    foreach (libxml_get_errors() as $error) {
        /* just ignore warnings */
        if ($error->level === LIBXML_ERR_WARNING) {
            continue;
        }
        fof_log(__FUNCTION__ . ': ' . $error->message);
    }
    libxml_clear_errors();
    libxml_use_internal_errors($old_xml_err);
    return $content_out;
}

 public function create($replaceHtml = false)
 {
     $this->_html = file_get_contents($this->getViewFile());
     // this is relative root directory
     $dom = new \DOMDocument();
     //        libxml_use_internal_errors(true);
     $dom->loadHTML($this->_html);
     //        libxml_clear_errors();
     if ($replaceHtml) {
         // use this if we ever need to get rid of doc types or html tags that we for some reason get wrapped in when calling loadHTML
         // remove doc type
         $dom->removeChild($dom->firstChild);
         //  remove <html><body></body></html>
         $dom->replaceChild($dom->firstChild->firstChild->firstChild, $dom->firstChild);
     }
     foreach ($this->getViewElements() as $viewElement) {
         $viewElement->replaceHtml($dom, $this->getViewElementValue());
     }
     $html = $dom->saveHTML();
     return $this->_html = trim($html);
 }

 public function loadHtmlNoCharset($htmlString = '')
 {
     $dom = new DOMDocument('1.0', 'UTF-8');
     $dom->preserveWhiteSpace = false;
     if (strlen($htmlString)) {
         libxml_use_internal_errors(true);
         $dom->loadHTML('<?xml encoding="UTF-8">' . $htmlString);
         // dirty fix
         foreach ($dom->childNodes as $item) {
             if ($item->nodeType == XML_PI_NODE) {
                 $dom->removeChild($item);
                 // remove hack
                 break;
             }
         }
         $dom->encoding = 'UTF-8';
         // insert proper
         libxml_clear_errors();
     }
     $this->loadDom($dom);
 }

 function process(&$article)
 {
     $owner_uid = $article["owner_uid"];
     if (strpos($article["link"], "mrlovenstein.com") !== FALSE) {
         if (strpos($article["plugin_data"], "af_comics,{$owner_uid}:") === FALSE) {
             $doc = new DOMDocument();
             @$doc->loadHTML($article["content"]);
             if ($doc) {
                 $xpath = new DOMXPath($doc);
                 $entries = $xpath->query('(//img[@alt])');
                 $basenode = false;
                 foreach ($entries as $entry) {
                     // get image
                     $basenode = $entry->parentNode;
                     // add linebreak
                     $linebreak = $doc->createElement("br");
                     $basenode->appendChild($linebreak);
                     // add text
                     $alt = $entry->getAttribute("alt");
                     $textnode = $doc->createTextNode($alt);
                     $basenode->appendChild($textnode);
                     break;
                 }
                 if ($basenode) {
                     $doc->removeChild($doc->firstChild);
                     $article["content"] = $doc->saveHTML();
                     $article["plugin_data"] = "af_comics,{$owner_uid}:" . $article["plugin_data"];
                 }
             }
         } else {
             if (isset($article["stored"]["content"])) {
                 $article["content"] = $article["stored"]["content"];
             }
         }
         return true;
     }
     return false;
 }