Diese Seite bei php.net
The PHP manual text and comments are covered by the Creative Commons Attribution 3.0 License © the PHP Documentation Group - Impressum - mail("TO:Reinhard Neidl",...)
(PHP 5)
DOMDocument::saveHTML — Dumps the internal document into a string using HTML formatting
Creates an HTML document from the DOM representation. This function is usually called after building a new dom document from scratch as in the example below.
Returns the HTML, or FALSE if an error occurred.
Beispiel #1 Saving a HTML tree into a string
<?php
$doc = new DOMDocument('1.0');
$root = $doc->createElement('html');
$root = $doc->appendChild($root);
$head = $doc->createElement('head');
$head = $root->appendChild($head);
$title = $doc->createElement('title');
$title = $head->appendChild($title);
$text = $doc->createTextNode('This is the title');
$text = $title->appendChild($text);
echo $doc->saveHTML();
?>
Another way to workaround the <script/> problem is putting a semicolon (;) inside the script element.
If you want a simpler way to get around the <script> tag problem try:
<?php
$script = $doc->createElement ('script');\
// Creating an empty text node forces <script></script>
$script->appendChild ($doc->createTextNode (''));
$head->appendChild ($script);
?>
To avoid script tags from being output as <script />, you can use the DOMDocumentFragment class:
<?php
$doc = new DOMDocument();
$doc -> loadXML($xmlstring);
$fragment = $doc->createDocumentFragment();
/* Append the script element to the fragment using raw XML strings (will be preserved in their raw form) and if succesful proceed to insert it in the DOM tree */
if($fragment->appendXML("<script type='text/javascript' src='$source'></script>") {
$xpath = new DOMXpath($doc);
$resultlist = $xpath->query("//*[local-name() = 'html']/*[local-name() = 'head']"); /* namespace-safe method to find all head elements which are childs of the html element, should only return 1 match */
foreach($resultlist as $headnode) // insert the script tag
$headnode->appendChild($fragment);
}
$doc->saveXML(); /* and our script tags will still be <script></script> */
?>
I am using this solution to prevent tags and the doctype from being added to the HTML string automatically:
<?php
$html = '<h1>Hello world!</h1>';
$html = '<div>' . $html . '</div>';
$doc = new DOMDocument;
$doc->loadHTML($html);
echo substr($doc->saveXML($doc->getElementsByTagName('div')->item(0)), 5, -6)
// Outputs: "<h1>Hello world!</h1>"
?>
This method, as of 5.2.6, will automatically add <html><body> and <!DOCTYPE> tags to the document if they are missing, without asking whether you want them. In my application, I needed to use the DOM methods to manipulate just a fragment of html, so these tags were rather unhelpful.
Here's a simple hack to remove them in case, like me, all you wanted to do was perform a few operations on an HTML fragment.
$html_fragment = preg_replace('/^<!DOCTYPE.+?>/', '', str_replace( array('<html>', '</html>', '<body>', '</body>'), array('', '', '', ''), $dom->saveHTML()));
<?php
function getDOMString($retNode) {
if (!$retNode) return null;
$retval = strtr($retNode-->ownerDocument->saveXML($retNode),
array(
'></area>' => ' />',
'></base>' => ' />',
'></basefont>' => ' />',
'></br>' => ' />',
'></col>' => ' />',
'></frame>' => ' />',
'></hr>' => ' />',
'></img>' => ' />',
'></input>' => ' />',
'></isindex>' => ' />',
'></link>' => ' />',
'></meta>' => ' />',
'></param>' => ' />',
'default:' => '',
// sometimes, you have to decode entities too...
'"' => '"',
'&' => '&',
''' => ''',
'<' => '<',
'>' => '>',
' ' => ' ',
'©' => '©',
'«' => '«',
'®' => '®',
'»' => '»',
'™' => '™'
));
return $retval;
}
?>
DOMDocument->saveXML() doesn't generate a proper XHTML format either.
There is a problem with "script" empty elements. For example:
This will be the code generated by saveXML, with an empty script tag.
<html>
<head>
<script type="text/JavaScript" src="myScript.js"/>
</head>
<body>
<p>I will not appear</p>
<script type="text/JavaScript">
alert("Not working");
</script>
</body>
</html>
I don't know if this is valid XHTML (W3C Validator doesn't complain), but both FF 2.0 and IE 6 will not render it properly. Both will use </script> as the closing tag for the first script causing js errors and ignoring in between elements.
You can post-process saveXML string in order to close empty tags with the following function:
<?php
function cerrarTag($tag, $xml){
$indice = 0;
while ($indice< strlen($xml)){
$pos = strpos($xml, "<$tag ", $indice);
if ($pos){
$posCierre = strpos($xml, ">", $pos);
if ($xml[$posCierre-1] == "/"){
$xml = substr_replace($xml, "></$tag>", $posCierre-1, 2);
}
$indice = $posCierre;
}
else break;
}
return $xml;
}
?>
At least script and select empty elements should be closed. This example shows how it can be used:
<?php
define("CABECERA_XHTML", '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">');
$xhtml = $docXML->saveXML($docXML->documentElement);
$xhtml = cerrarTag("script", $xhtml);
$xhtml = cerrarTag("select", $xhtml);
$xhtml = CABECERA_XHTML."\n".$xhtml;
echo $xhtml;
?>
If created your DOMDocument object using loadHTML() (where the source is from another site) and want to pass your changes back to the browser you should make sure the HTTP Content-Type header matches your meta content-type tags value because modern browsers seem to ignore the meta tag and trust just the HTTP header. For example if you're reading an ISO-8859-1 document and your web server is claiming UTF-8 you need to correct it using the header() function.
<?php
header('Content-Type: text/html; charset=iso-8859-1');
?>
XHTML:
If the output is XHTML use the function saveXML().
Output example for saveHTML:
<select name="pet" size="3" multiple>
<option selected>mouse</option>
<option>bird</option>
<option>cat</option>
</select>
XHTML conform output using saveXML:
<select name="pet" size="3" multiple="multiple">
<option selected="selected">mouse</option>
<option>bird</option>
<option>cat</option>
</select>
<?php
// Using DOM to fix sloppy HTML.
// An example by Tyson Clugg <tyson@clugg.net>
//
// vim: syntax=php expandtab tabstop=2
function tidyHTML($buffer)
{
// load our document into a DOM object
$dom = @DOMDocument::loadHTML($buffer);
// we want nice output
$dom->formatOutput = true;
return($dom->saveHTML());
}
// start output buffering, using our nice
// callback funtion to format the output.
ob_start("tidyHTML");
?>
<html>
<p>It's like comparing apples to oranges.
</html>
<?php
// this will be called implicitly, but we'll
// call it manually to illustrate the point.
ob_end_flush();
?>
The above code takes out sloppy HTML:
<html>
<p>It's like comparing apples to oranges.
</html>
And cleans it up to the following:
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
<html><body><p>It's like comparing apples to oranges.
</p></body></html>
Diese Seite bei php.net