public function __construct($sValue)
{
if (Sylma::read('dom/encoding/check') && !mb_check_encoding($sValue, 'UTF-8')) {
//$sValue = utf8_encode($sContent); // TODO , result not always in utf-8
$this->getParent()->throwException('Bad encoding');
}
parent::__construct($sValue);
}
/**
* Gets line number for where the node is defined.
*
* The method `DOMNode::getLineNo()` does not return the correct line number for text nodes. So this method
* can return either the original line number returned by libxml or the fixed and correct line number
* created by the `XMLDocument::fixLineNumbers()` method.
*
* @return int Either the original line number returned by libxml or the fixed and correct line number
* created by the `XMLDocument::fixLineNumbers()` method.
*/
public function getLineNo()
{
$container = $this->ownerDocument->getLineNumberContainer();
if ($container && $container->contains($this)) {
return $container->offsetGet($this);
}
return parent::getLineNo();
}
function __toString()
{
// Skip empty text frames
// if ( $this->is_text_node() &&
// preg_replace("/\s/", "", $this->_node->data) === "" )
// return "";
$str = "<b>" . $this->_node->nodeName . ":</b><br/>";
//$str .= spl_object_hash($this->_node) . "<br/>";
$str .= "Id: " . $this->get_id() . "<br/>";
$str .= "Class: " . get_class($this) . "<br/>";
if ($this->is_text_node()) {
$tmp = htmlspecialchars($this->_node->nodeValue);
$str .= "<pre>'" . mb_substr($tmp, 0, 70) . (mb_strlen($tmp) > 70 ? "..." : "") . "'</pre>";
} elseif ($css_class = $this->_node->getAttribute("class")) {
$str .= "CSS class: '{$css_class}'<br/>";
}
if ($this->_parent) {
$str .= "\nParent:" . $this->_parent->_node->nodeName . " (" . spl_object_hash($this->_parent->_node) . ") " . "<br/>";
}
if ($this->_prev_sibling) {
$str .= "Prev: " . $this->_prev_sibling->_node->nodeName . " (" . spl_object_hash($this->_prev_sibling->_node) . ") " . "<br/>";
}
if ($this->_next_sibling) {
$str .= "Next: " . $this->_next_sibling->_node->nodeName . " (" . spl_object_hash($this->_next_sibling->_node) . ") " . "<br/>";
}
$d = $this->get_decorator();
while ($d && $d != $d->get_decorator()) {
$str .= "Decorator: " . get_class($d) . "<br/>";
$d = $d->get_decorator();
}
$str .= "Position: " . pre_r($this->_position, true);
$str .= "\nContaining block: " . pre_r($this->_containing_block, true);
$str .= "\nMargin width: " . pre_r($this->get_margin_width(), true);
$str .= "\nMargin height: " . pre_r($this->get_margin_height(), true);
$str .= "\nStyle: <pre>" . $this->_style->__toString() . "</pre>";
if ($this->_decorator instanceof Block_Frame_Decorator) {
$str .= "Lines:<pre>";
foreach ($this->_decorator->get_line_boxes() as $line) {
foreach ($line->get_frames() as $frame) {
if ($frame instanceof Text_Frame_Decorator) {
$str .= "\ntext: ";
$str .= "'" . htmlspecialchars($frame->get_text()) . "'";
} else {
$str .= "\nBlock: " . $frame->get_node()->nodeName . " (" . spl_object_hash($frame->get_node()) . ")";
}
}
$str .= "\ny => " . $line->y . "\n" . "w => " . $line->w . "\n" . "h => " . $line->h . "\n" . "left => " . $line->left . "\n" . "right => " . $line->right . "\n";
}
$str .= "</pre>";
}
$str .= "\n";
if (php_sapi_name() === "cli") {
$str = strip_tags(str_replace(array("<br/>", "<b>", "</b>"), array("\n", "", ""), $str));
}
return $str;
}
function set_id($id)
{
$this->_id = $id;
// We can only set attributes of DOMElement objects (nodeType == 1).
// Since these are the only objects that we can assign CSS rules to,
// this shortcoming is okay.
if ($this->_node->nodeType == XML_ELEMENT_NODE) {
$this->_node->setAttribute("frame_id", $id);
}
}
function set_content($text)
{
// the following replace has been added to conform with PHP4.
// A set_content("&") brought a get_content() = "&" there,
// whereas PHP5 gives a get_content() = "&"
$text = str_replace("<", "<", $text);
$text = str_replace(">", ">", $text);
$text = str_replace("&", "&", $text);
$text_node = new DOMText();
$text_node->appendData($text);
if (is_object($this->myDOMNode->firstChild)) {
$this->myDOMNode->replaceChild($text_node, $this->myDOMNode->firstChild);
} else {
$this->myDOMNode->appendChild($text_node);
}
}
/**
* Highlight text in text node
*
* @param DOMText $node
* @param array $wordsToHighlight
* @param callback $callback Callback method, used to transform (highlighting) text.
* @param array $params Array of additionall callback parameters (first non-optional parameter is a text to transform)
* @throws \Zend\Search\Lucene\Exception\RuntimeException
*/
protected function _highlightTextNode(\DOMText $node, $wordsToHighlight, $callback, $params)
{
$analyzer = Analyzer\Analyzer::getDefault();
$analyzer->setInput($node->nodeValue, 'UTF-8');
$matchedTokens = array();
while (($token = $analyzer->nextToken()) !== null) {
if (isset($wordsToHighlight[$token->getTermText()])) {
$matchedTokens[] = $token;
}
}
if (count($matchedTokens) == 0) {
return;
}
$matchedTokens = array_reverse($matchedTokens);
foreach ($matchedTokens as $token) {
// Cut text after matched token
$node->splitText($token->getEndOffset());
// Cut matched node
$matchedWordNode = $node->splitText($token->getStartOffset());
// Retrieve HTML string representation for highlihted word
$fullCallbackparamsList = $params;
array_unshift($fullCallbackparamsList, $matchedWordNode->nodeValue);
$highlightedWordNodeSetHTML = call_user_func_array($callback, $fullCallbackparamsList);
// Transform HTML string to a DOM representation and automatically transform retrieved string
// into valid XHTML (It's automatically done by loadHTML() method)
$highlightedWordNodeSetDomDocument = new \DOMDocument('1.0', 'UTF-8');
$success = @$highlightedWordNodeSetDomDocument->loadHTML('<html><head><meta http-equiv="Content-type" content="text/html; charset=UTF-8"/></head><body>' . $highlightedWordNodeSetHTML . '</body></html>');
if (!$success) {
throw new RuntimeException("Error occured while loading highlighted text fragment: '{$highlightedWordNodeSetHTML}'.");
}
$highlightedWordNodeSetXpath = new \DOMXPath($highlightedWordNodeSetDomDocument);
$highlightedWordNodeSet = $highlightedWordNodeSetXpath->query('/html/body')->item(0)->childNodes;
for ($count = 0; $count < $highlightedWordNodeSet->length; $count++) {
$nodeToImport = $highlightedWordNodeSet->item($count);
$node->parentNode->insertBefore($this->_doc->importNode($nodeToImport, true), $matchedWordNode);
}
$node->parentNode->removeChild($matchedWordNode);
}
}
/**
* Highlight text in text node
*
* @param DOMText $node
* @param array $wordsToHighlight
* @param string $color
*/
public function _highlightTextNode(DOMText $node, $wordsToHighlight, $color)
{
$analyzer = Zend_Search_Lucene_Analysis_Analyzer::getDefault();
$analyzer->setInput($node->nodeValue, $this->_doc->encoding);
$matchedTokens = array();
while (($token = $analyzer->nextToken()) !== null) {
if (isset($wordsToHighlight[$token->getTermText()])) {
$matchedTokens[] = $token;
}
}
if (count($matchedTokens) == 0) {
return;
}
$matchedTokens = array_reverse($matchedTokens);
foreach ($matchedTokens as $token) {
// Cut text after matched token
$node->splitText($token->getEndOffset());
// Cut matched node
$matchedWordNode = $node->splitText($token->getStartOffset());
$highlightedNode = $this->_doc->createElement('b', $matchedWordNode->nodeValue);
$highlightedNode->setAttribute('style', 'color:black;background-color:' . $color);
$node->parentNode->replaceChild($highlightedNode, $matchedWordNode);
}
}
/**
* Highlight text in text node
*
* @param DOMText $node
* @param array $wordsToHighlight
* @param callback $callback Callback method, used to transform (highlighting) text.
* @param array $params Array of additionall callback parameters (first non-optional parameter is a text to transform)
* @throws Zend_Search_Lucene_Exception
*/
protected function _highlightTextNode(DOMText $node, $wordsToHighlight, $callback, $params)
{
/** Zend_Search_Lucene_Analysis_Analyzer */
require_once 'Zend/Search/Lucene/Analysis/Analyzer.php';
$analyzer = Zend_Search_Lucene_Analysis_Analyzer::getDefault();
$analyzer->setInput($node->nodeValue, 'UTF-8');
$matchedTokens = array();
while (($token = $analyzer->nextToken()) !== null) {
if (isset($wordsToHighlight[$token->getTermText()])) {
$matchedTokens[] = $token;
}
}
if (count($matchedTokens) == 0) {
return;
}
$matchedTokens = array_reverse($matchedTokens);
foreach ($matchedTokens as $token) {
// Cut text after matched token
$node->splitText($token->getEndOffset());
// Cut matched node
$matchedWordNode = $node->splitText($token->getStartOffset());
// Retrieve HTML string representation for highlihted word
$fullCallbackparamsList = $params;
array_unshift($fullCallbackparamsList, $matchedWordNode->nodeValue);
$highlightedWordNodeSetHtml = call_user_func_array($callback, $fullCallbackparamsList);
// Transform HTML string to a DOM representation and automatically transform retrieved string
// into valid XHTML (It's automatically done by loadHTML() method)
$highlightedWordNodeSetDomDocument = new DOMDocument('1.0', 'UTF-8');
$success = @$highlightedWordNodeSetDomDocument->loadHTML($highlightedWordNodeSetHtml);
if (!$success) {
require_once 'Zend/Search/Lucene/Exception.php';
throw new Zend_Search_Lucene_Exception("Error occured while loading highlighted text fragment: '{$highlightedWordNodeSetHtml}'.");
}
$highlightedWordNodeSetXpath = new DOMXPath($highlightedWordNodeSetDomDocument);
$highlightedWordNodeSet = $highlightedWordNodeSetXpath->query('/html/body')->item(0)->childNodes;
for ($count = 0; $count < $highlightedWordNodeSet->length; $count++) {
$nodeToImport = $highlightedWordNodeSet->item($count);
$node->parentNode->insertBefore($this->_doc->importNode($nodeToImport, true), $matchedWordNode);
}
$node->parentNode->removeChild($matchedWordNode);
}
}
$comment = new DOMComment('Testing character data and extending nodes');
$charnode->appendChild($comment);
echo "Comment Length: " . $comment->length . "\n";
$comment->data = 'Updated comment';
echo "New Comment Length: " . $comment->length . "\n";
echo "New Comment Data: " . $comment->data . "\n";
/* DOMCDataSection */
$cdata = new DOMCDataSection('Chars: <>&"');
$charnode->appendChild($cdata);
echo "Substring: " . $cdata->substringData(7, 4) . "\n";
$cdata->replaceData(10, 1, "'");
echo "New Substring: " . $cdata->substringData(7, 4) . "\n";
/* DOMCharacterData using DOMComment */
$comment = new DOMComment('instructions');
echo "Comment Value: " . $comment->data . "\n";
$comment->data = 'some more instructions';
echo "New Comment Value: " . $comment->data . "\n";
$comment->insertData(10, 'pi ');
$comment->replaceData(18, 5, 'i');
$comment->insertData(20, 'g');
$comment->deleteData(13, 2);
$comment->deleteData(10, 3);
$comment->insertData(10, 'comment ');
echo "Updated Comment Value: " . $comment->data . "\n";
/* DOMText */
$text = new DOMText('some text characters');
echo "Whole Text: " . $text->wholeText . "\n";
$text2 = $text->splitText(9);
echo "Split text: " . $text2->wholeText . "\n";
$text3 = $text2->splitText(1);
echo "Is Whitespace?: " . ($text2->isElementContentWhitespace() ? 'YES' : 'NO');
/**
* This method attempts to return a better 'innerText' representation than DOMNode::textContent
*
* @param DOMElement|DOMText $el
* @param bool $implied when parsing for implied name for h-*, rules may be slightly different
* @see: https://github.com/glennjones/microformat-shiv/blob/dev/lib/text.js
*/
public function innerText($el, $implied = false)
{
$out = '';
$blockLevelTags = array('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p', 'hr', 'pre', 'table', 'address', 'article', 'aside', 'blockquote', 'caption', 'col', 'colgroup', 'dd', 'div', 'dt', 'dir', 'fieldset', 'figcaption', 'figure', 'footer', 'form', 'header', 'hgroup', 'hr', 'li', 'map', 'menu', 'nav', 'optgroup', 'option', 'section', 'tbody', 'testarea', 'tfoot', 'th', 'thead', 'tr', 'td', 'ul', 'ol', 'dl', 'details');
$excludeTags = array('noframe', 'noscript', 'script', 'style', 'frames', 'frameset');
// PHP DOMDocument doesn’t correctly handle whitespace around elements it doesn’t recognise.
$unsupportedTags = array('data');
if (isset($el->tagName)) {
if (in_array(strtolower($el->tagName), $excludeTags)) {
return $out;
} else {
if ($el->tagName == 'img') {
if ($el->getAttribute('alt') !== '') {
return $el->getAttribute('alt');
} else {
if (!$implied && $el->getAttribute('src') !== '') {
return $this->resolveUrl($el->getAttribute('src'));
}
}
} else {
if ($el->tagName == 'area' and $el->getAttribute('alt') !== '') {
return $el->getAttribute('alt');
} else {
if ($el->tagName == 'abbr' and $el->getAttribute('title') !== '') {
return $el->getAttribute('title');
}
}
}
}
}
// if node is a text node get its text
if (isset($el->nodeType) && $el->nodeType === 3) {
$out .= $el->textContent;
}
// get the text of the child nodes
if ($el->childNodes && $el->childNodes->length > 0) {
for ($j = 0; $j < $el->childNodes->length; $j++) {
$text = $this->innerText($el->childNodes->item($j), $implied);
if (!is_null($text)) {
$out .= $text;
}
}
}
if (isset($el->tagName)) {
// if its a block level tag add an additional space at the end
if (in_array(strtolower($el->tagName), $blockLevelTags)) {
$out .= ' ';
} elseif ($implied and in_array(strtolower($el->tagName), $unsupportedTags)) {
$out .= ' ';
} else {
if (strtolower($el->tagName) == 'br') {
// else if its a br, replace with newline
$out .= "\n";
}
}
}
return $out === '' ? NULL : $out;
}
