// If a $doc is handed over (an ExtendedDocument or a derived class), load the content into that document.
// returns an associative array with the following elements: 'html', 'head', 'title', 'body'
//
+ // public function loadHTML5($source) {
+ // A version of loadHTML() - see DOMDocument documentation - that is made for loading HTML5 and not emitting warnings/errors for unknown elements.
+ // returns true on success, false otherwise, just like loadHTML5.
+ //
// public function appendElement($name, [$value])
// appends a DOMDocument::createElement() as a child of this document (see there for params)
// returns the new child
static function initHTML5($doc = null) {
if (is_null($doc)) { $doc = new ExtendedDocument(); }
- $doc->loadHTML('<?xml version="1.0" encoding="utf-8"?><!DOCTYPE html><html></html>'); // this seems to be the only way to get the DOCTYPE set properly.
+ $doc->loadHTML5('<?xml version="1.0" encoding="utf-8"?><!DOCTYPE html><html></html>'); // this seems to be the only way to get the DOCTYPE set properly.
// Created basic HTML document structure.
$root = $doc->getElementsByTagName('html')->item(0);
'body' => $body);
}
+ public function loadHTML5($source) {
+ // Do our own handling of DOMDocument error reporting so we can ignore "unknown tags" which are usually fine in HTML5.
+ libxml_use_internal_errors(true);
+ if (!preg_match('/^\s*<?xml /', $source)) {
+ // Add an XML declaration to force DOMDocument into UTF-8 mode.
+ $source = '<?xml version="1.0" encoding="utf-8"?>'."\n".$source;
+ }
+ $result = $this->loadHTML($source);
+ // Handle DOMDocument loading errors, throw away warnings on unknown tags as HTML5 allows all kinds.
+ $errseverity = array(LIBXML_ERR_WARNING => 'Warning', LIBXML_ERR_ERROR => 'Error', LIBXML_ERR_FATAL => 'Fatal');
+ foreach (libxml_get_errors() as $error) {
+ // $error is a libXMLError, see https://www.php.net/manual/en/class.libxmlerror.php
+ // See http://www.xmlsoft.org/html/libxml-xmlerror.html#xmlParserErrors for error numbers
+ if ($error->code != 801) { // XML_HTML_UNKNOWN_TAG gets no output, should not exist for HTML5.
+ trigger_error($errseverity[$error->level].' loading HTML5: '.$error->message.' (code '.$error->code.'), line: '.$error->line, E_USER_WARNING);
+ }
+ }
+ libxml_clear_errors();
+ libxml_use_internal_errors(false);
+ return $result;
+ }
+
public function appendElement($name, $value = '') {
return $this->appendChild($this->createElement($name, $value));
}
public function appendHTMLMarkup($htmldata, $parentNode = null) {
if (is_null($parentNode)) { $parentNode =& $this; }
- // Use loadHTML() to parse and turn the markup into proper HTML.
+ // Use loadHTML5() to parse and turn the markup into proper HTML.
$tmpdoc = new ExtendedDocument;
// The XML line is needed to tell the parser that we need UTF-8 parsing.
- $tmpdoc->loadHTML('<?xml version="1.0" encoding="utf-8"?><!DOCTYPE html><html><body>'.$htmldata.'</body></html>');
+ $tmpdoc->loadHTML5('<?xml version="1.0" encoding="utf-8"?><!DOCTYPE html><html><body>'.$htmldata.'</body></html>');
foreach ($tmpdoc->getElementsByTagName('body')->item(0)->childNodes as $child) {
$parentNode->appendChild($this->importNode($child, true));
}
projects / php-utility-classes.git / commitdiff