protected function get_permalink_as_dom($path)
{
ob_start();
$this->go_to($path);
$this->load_template();
$output = ob_get_clean();
$html5 = new HTML5();
$dom = $html5->loadHTML($output);
return $dom;
}
/**
* Stores a view output in the elements.
*/
function storeViewPreview($output)
{
$html5 = new HTML5();
$htmlDom = $html5->loadHTML('<html><body>' . $output . '</body></html>');
if ($htmlDom) {
// It's much easier to work with simplexml than DOM, luckily enough
// we can just simply import our DOM tree.
$this->elements = simplexml_import_dom($htmlDom);
}
}
public function process($content)
{
if (strlen($content) == 0) {
return '';
}
$fragment = $this->html5->loadHTMLFragment($content);
$fragment = $this->processModifiers($fragment);
return $this->html5->saveHTML($fragment);
}
/**
* Parse HTML5 and convert to Dom document
*
* @param $string
*
* @return false|\DomDocument
*/
public function parseFromString($string)
{
$dom = $this->html5->loadHTML($string);
if (is_a($dom, 'DOMDocument')) {
return $dom;
} else {
return false;
}
}
/**
* Fix markup
*
* @param string $markup
* @param int $topLevel
* @param int $depth
* @return string Markup with added IDs
* @throws RuntimeException
*/
public function fix($markup, $topLevel = 1, $depth = 6)
{
if (!$this->isFullHtmlDocument($markup)) {
$partialID = 'toc_generator_' . mt_rand(1000, 4000);
$markup = sprintf("<body id='%s'>%s</body>", $partialID, $markup);
}
$domDocument = $this->htmlParser->loadHTML($markup);
$domDocument->preserveWhiteSpace = true;
// do not clobber whitespace
$sluggifier = new UniqueSluggifier();
/** @var \DOMElement $node */
foreach ($this->traverseHeaderTags($domDocument, $topLevel, $depth) as $node) {
if ($node->getAttribute('id')) {
continue;
}
$node->setAttribute('id', $sluggifier->slugify($node->getAttribute('title') ?: $node->textContent));
}
return $this->htmlParser->saveHTML(isset($partialID) ? $domDocument->getElementById($partialID)->childNodes : $domDocument);
}
/**
* Get Menu
*
* Returns a KNP Menu object, which can be traversed or rendered
*
* @param string $markup Content to get items fro $this->getItems($markup, $topLevel, $depth)m
* @param int $topLevel Top Header (1 through 6)
* @param int $depth Depth (1 through 6)
* @return ItemInterface KNP Menu
*/
public function getMenu($markup, $topLevel = 1, $depth = 6)
{
// Setup an empty menu object
$menu = $this->menuFactory->createItem('TOC');
// Empty? Do nothing.
if (trim($markup) == '') {
return [];
}
// Parse HTML
$tagsToMatch = $this->determineHeaderTags($topLevel, $depth);
// Initial settings
$lastElem = $menu;
// Do it...
$domDocument = $this->domParser->loadHTML($markup);
foreach ($this->traverseHeaderTags($domDocument, $topLevel, $depth) as $node) {
// Skip items without IDs
if (!$node->hasAttribute('id')) {
continue;
}
// Get the TagName and the level
$tagName = $node->tagName;
$level = array_search(strtolower($tagName), $tagsToMatch) + 1;
// Determine parent item which to add child
if ($level == 1) {
$parent = $menu;
} elseif ($level == $lastElem->getLevel()) {
$parent = $lastElem->getParent();
} elseif ($level > $lastElem->getLevel()) {
$parent = $lastElem;
for ($i = $lastElem->getLevel(); $i < $level - 1; $i++) {
$parent = $parent->addChild('');
}
} else {
//if ($level < $lastElem->getLevel())
$parent = $lastElem->getParent();
while ($parent->getLevel() > $level - 1) {
$parent = $parent->getParent();
}
}
$lastElem = $parent->addChild($node->getAttribute('title') ?: $node->textContent, ['uri' => '#' . $node->getAttribute('id')]);
}
return $menu;
}
/**
* Load a HTML5 file and copy it into a FluentDOM\Document
*
* @codeCoverageIgnore
*
* @see Loadable::load
* @param string $source
* @param string $contentType
* @param array $options
* @return Document|NULL
*/
public function load($source, $contentType, array $options = [])
{
if ($this->supports($contentType)) {
$html5 = new HTML5Support();
if ($this->startsWith($source, '<')) {
$dom = $html5->loadHTML($source);
} else {
$dom = $html5->loadHTMLFile($source);
}
if (!$dom instanceof Document) {
$fd = new Document();
if ($dom->documentElement instanceof \DOMElement) {
$fd->appendChild($fd->importNode($dom->documentElement, TRUE));
}
$dom = $fd;
}
$dom->registerNamespace('html', 'http://www.w3.org/1999/xhtml');
return $dom;
}
return NULL;
}
/**
* {@inheritdoc}
*/
public function toHtml(\DOMDocument $document, $encoding = 'UTF-8')
{
return $this->html5Parser->saveHTML($document);
}
<?php
require "vendor/autoload.php";
use Masterminds\HTML5;
$html = <<<'HERE'
<html>
<head>
<title>TEST</title>
<script language="javascript">
if (2 > 1) { alert("Math wins."); }
</script>
</head>
<body id='foo'>
<!-- This space intentionally left blank. -->
<section class="section-a pretty" id="bar1">
<h1>Hello World</h1><p>This is a test of the HTML5 parser.</p>
<hr>
& Nobody nowhere.
</section>
<test xmlns:foo="http://example.com/foo">TEST</test>
<![CDATA[Because we can.]]>
©
</body></html>
HERE;
$html5 = new HTML5();
$dom = $html5->loadHTML($html);
print "Converting to HTML 5\n";
$html5->save($dom, fopen("php://stdin", 'w'));
/**
* Write the document to HTML5.
*
* This works the same as the other write* functions, but it encodes the output
* as HTML5 with UTF-8.
* @see html5()
* @see innerHTML5()
* @throws Exception
* In the event that a file cannot be written, an Exception will be thrown.
*/
public function writeHTML5($path = NULL)
{
$html5 = new HTML5();
if ($path == NULL) {
// Print the document to stdout.
print $html5->saveHTML($this->document);
return;
}
$html5->save($this->document, $path);
}
/**
* Parse HTML5 documents.
*
* This uses HTML5-PHP to parse the document. In actuality, this parser does
* a fine job with pre-HTML5 documents in most cases, though really old HTML
* (like 2.0) may have some substantial quirks.
*
* <b>Supported Options</b>
* Any options supported by HTML5-PHP are allowed here. Additionally, the
* following options have meaning to QueryPath.
* - QueryPath_class
*
*
* @param mixed $source
* A document as an HTML string, or a path/URL. For compatibility with
* existing functions, a DOMDocument, SimpleXMLElement, DOMNode or array
* of DOMNodes will be passed through as well. However, these types are not
* validated in any way.
*
* @param string $selector
* A CSS3 selector.
*
* @param array $options
* An associative array of options, which is passed on into HTML5-PHP. Note
* that the standard QueryPath options may be ignored for this function,
* since it uses a different parser.
*
* @return QueryPath
*/
public static function withHTML5($source = NULL, $selector = NULL, $options = array())
{
$qpClass = isset($options['QueryPath_class']) ? $options['QueryPath_class'] : '\\QueryPath\\DOMQuery';
if (is_string($source)) {
$html5 = new HTML5();
if (strpos($source, '<') !== FALSE && strpos($source, '>') !== FALSE) {
$source = $html5->loadHTML($source);
} else {
$source = $html5->load($source);
}
}
$qp = new $qpClass($source, $selector, $options);
return $qp;
}
/**
* Return body as string
*
* @param \DOMDocument $dom
*
* @return mixed
*/
public static function to_string(\DOMDocument $dom)
{
$html5 = new HTML5();
preg_match('/<body>(.*)<\\/body>/s', $html5->saveHTML($dom), $match);
return $match[1];
}
public function asString()
{
$html5 = new HTML5Support($this->_options);
return (string) $html5->saveHTML($this->_document);
}
foreach ($contentsNodes->getElementsByTagName('span') as $span) {
if ($span->getAttribute('class') == 'page-name') {
$span->appendChild($contentsNodes->createTextNode($requestedPageName));
}
}
// script tags may appear above the script block - can't have that!
$scriptTags = [];
foreach ($contentsNodes->getElementsByTagName('script') as $script) {
$scriptTags[] = $script->cloneNode(true);
$script->parentNode->removeChild($script);
}
/*************************/
/* Load the outer page */
/*************************/
$index = file_get_contents("../index.html");
$html5Parser = new HTML5();
$completePage = $html5Parser->loadHTML($index);
/*************************/
/* Merge */
/*************************/
// find the destination
$destination = $completePage->getElementById('cms-less-destination');
$destination->setAttribute("data-cms-less-preloaded", $originalPageName);
// empty the destination
foreach ($destination->childNodes as $node) {
$destination->removeChild($node);
}
// load the page content
$childNodes = $contentsNodes->childNodes;
foreach ($childNodes as $node) {
$importedNode = $completePage->importNode($node, true);
/**
* Test alternate constructors.
* @group basic
*/
public function testDOMQueryHtmlConstructors()
{
$qp = htmlqp(\QueryPath::HTML_STUB);
$this->assertEquals(1, count($qp->get()));
$this->assertTrue($qp->get(0) instanceof \DOMNode);
// Bad BR tag.
$borken = '<html><head></head><body><br></body></html>';
$qp = htmlqp($borken);
$this->assertEquals(1, count($qp->get()));
$this->assertTrue($qp->get(0) instanceof \DOMNode);
// XHTML Faker
$borken = '<?xml version="1.0"?><html><head></head><body><br></body></html>';
$qp = htmlqp($borken);
$this->assertEquals(1, count($qp->get()));
$this->assertTrue($qp->get(0) instanceof \DOMNode);
// HTML in a file that looks like XML.
$qp = htmlqp(HTML_IN_XML_FILE);
$this->assertEquals(1, count($qp->get()));
$this->assertTrue($qp->get(0) instanceof \DOMNode);
// HTML5
$html5 = new \Masterminds\HTML5();
$dom = $html5->loadHTML(\QueryPath::HTML_STUB);
qp($dom, 'html');
// Stripping #13 (CR) from HTML.
$borken = '<html><head></head><body><p>' . chr(13) . '</p><div id="after"/></body></html>';
$this->assertFalse(strpos(htmlqp($borken)->html(), ' '), 'Test that CRs are not encoded.');
// Regression for #58: Make sure we aren't getting encoded.
$borken = '<html><head><style>
.BlueText {
color:red;
}</style><body></body></html>';
$this->assertFalse(strpos(htmlqp($borken)->html(), ' '), 'Test that LF is not encoded.');
// Low ASCII in a file
$borken = '<html><head></head><body><p>' . chr(27) . '</p><div id="after"/></body></html>';
$this->assertEquals(1, htmlqp($borken, '#after')->size());
}
