Warning: DOMDocument::loadHTML(): htmlParseEntityRef: expecting ';' in Entity,

Question

$html = file_get_contents("http://www.somesite.com/");

$dom = new DOMDocument();
$dom->loadHTML($html);

echo $dom;

throws

Warning: DOMDocument::loadHTML(): htmlParseEntityRef: expecting ';' in Entity,
Catchable fatal error: Object of class DOMDocument could not be converted to string in test.php on line 10

Answer 1

To evaporate the warning, you can use libxml_use_internal_errors(true)

// create new DOMDocument
$document = new \DOMDocument('1.0', 'UTF-8');

// set error level
$internalErrors = libxml_use_internal_errors(true);

// load HTML
$document->loadHTML($html);

// Restore error level
libxml_use_internal_errors($internalErrors);

Answer 2

I would bet that if you looked at the source of http://www.somesite.com/ you would find special characters that haven't been converted to HTML. Maybe something like this:

<a href="/script.php?foo=bar&hello=world">link</a>

Should be

<a href="/script.php?foo=bar&amp;hello=world">link</a>

Answer 3

$dom->@loadHTML($html);

This is incorrect, use this instead:

@$dom->loadHTML($html);

Answer 4

There are 2 errors: the second is because $dom is no string but an object and thus cannot be "echoed". The first error is a warning from loadHTML, caused by invalid syntax of the html document to load (probably an & (ampersand) used as parameter separator and not masked as entity with &).

You ignore and supress this error message (not the error, just the message!) by calling the function with the error control operator "@" (http://www.php.net/manual/en/language.operators.errorcontrol.php )

@$dom->loadHTML($html);

Answer 5

The reason for your fatal error is DOMDocument does not have a __toString() method and thus can not be echo'ed.

You're probably looking for

echo $dom->saveHTML();

Answer 6

Regardless of the echo (which would need to be replaced with print_r or var_dump), if an exception is thrown the object should stay empty:

DOMNodeList Object
(
)

Solution

1. Set recover to true, and strictErrorChecking to false

   $content = file_get_contents($url);
   
   $doc = new DOMDocument();
   $doc->recover = true;
   $doc->strictErrorChecking = false;
   $doc->loadHTML($content);
   

2. Use php's entity-encoding on the markup's contents, which is a most common error source.

Answer 7

replace the simple

$dom->loadHTML($html);

with the more robust ...

libxml_use_internal_errors(true);

if (!$DOM->loadHTML($page))
    {
        $errors="";
        foreach (libxml_get_errors() as $error)  {
            $errors.=$error->message."<br/>";
        }
        libxml_clear_errors();
        print "libxml errors:<br>$errors";
        return;
    }