First getElementsByTagName() returns all elements in HTML (Strange behaviour)

1 Views

I am using PHP to parse HTML provided to me by WordPress.

This is a post’s PHP returned my WordPress:

Read More
<p>Test</p> 
<p>
    <img class="alignnone size-thumbnail wp-image-39" src="img.png"/>
</p> 
<p>Ok.</p>

This is my parsing function (with debugging left in): 

function get_parsed_blog_post()
{
    $html = ob_wp_content(false);

    print_r(htmlspecialchars($html));
    echo '<hr/><hr/><hr/>';

    $parse = new DOMDocument();
    $parse->loadHTML($html, LIBXML_HTML_NOIMPLIED | LIBXML_HTML_NODEFDTD);

    $xpath = new DOMXpath($parse);
    $ps = $xpath->query('//p');

    foreach ($ps as $p) 
    {
        $imgs = $p->getElementsByTagName('img');

        print($imgs->length);
        echo '<br/>';

        if ($imgs->length > 0)
        {
            $p->setAttribute('class', 'image-content');

            foreach ($imgs as $img)
            {
                $img->removeAttribute('class');
            }
        }        
    }

    $htmlFinal = $parse->saveHTML();

    print_r(htmlspecialchars($htmlFinal));
    echo '<hr/><hr/><hr/>';

    return $htmlFinal;
}

The purpose of this code is to remove the classes WordPress adds to the <img>s, and to set any <p> that contains an image to be a class of image-content.

And this returns: 

1
1
0
<p class="image-content">Test
<p class="image-content">
    <img src="img.png">
</p>
<p>Ok.</p></p>

Somehow, it has wrapped the first occurrence of <p> around my entire parsed post, causing the first <p> to have the image-content class incorrectly applied. Why is this happening? How do I stop it?

Post Views: 1

Related posts

1 comment

  1. METHOD 1

    As to use exactly your code, I have done some changes to make it working.

    If you will print out each $p you will be able to see first element will contain all your HTML. The simplest solution is to add a blank <p> before your HTML and skip it when foreach.

    function get_parsed_blog_post()
{
    $page_content_html = ob_wp_content(false);
    $html = "<p></p>".$page_content_html;
    print_r(htmlspecialchars($html));
    echo '<hr/><hr/><hr/>';

    $parse = new DOMDocument();
    $parse->loadHTML($html, LIBXML_HTML_NOIMPLIED | LIBXML_HTML_NODEFDTD);

    $xpath = new DOMXpath($parse);
    $ps = $xpath->query('//p');
    $i = 0;
    foreach ($ps as $p) 
    {
        if($i != 0) {
            $imgs = $p->getElementsByTagName('img');

            print($imgs->length);
            echo '<br/>';

            if ($imgs->length > 0)
            {
                $p->setAttribute('class', 'image-content');

                foreach ($imgs as $img)
                {
                    $img->removeAttribute('class');
                }
            }
        }
        $i++;
    }

    $htmlFinal = $parse->saveHTML();

    print_r(htmlspecialchars($htmlFinal));             
    echo '<hr/><hr/><hr/>';

    return $htmlFinal;
}

    Total execution time in seconds: 0.00034999847412109

    METHOD 2

    The problem was caused by LIBXML_HTML_NOIMPLIED | LIBXML_HTML_NODEFDTD (which is making first <p> as a parent too), but you can remove document tags without this. So, you can do it as here:

    function get_parsed_blog_post()
{
$page_content_html = ob_wp_content(false);
$doc = new DOMDocument();
$doc->loadHTML($page_content_html);
foreach($doc->getElementsByTagName('p') as $paragraph) {
    $imgs = $paragraph->getElementsByTagName('img');
    if ($imgs->length > 0)
    {
        $paragraph->setAttribute('class', 'image-content');

        foreach ($imgs as $img)
        {
            $img->removeAttribute('class');
        }
    }        
}


/* REMOVING DOCTYPE, HTML AND BODY TAGS */

// Removing DOCTYPE
$doc->removeChild($doc->doctype);

// Removing HTML tag
$doc->replaceChild($doc->firstChild->firstChild, $doc->firstChild);

// Removing Body Tag
$html = $doc->getElementsByTagName("body")->item(0);
$fragment = $doc->createDocumentFragment();
while ($html->childNodes->length > 0) {
    $fragment->appendChild($html->childNodes->item(0));
}
$html->parentNode->replaceChild($fragment, $html);

$htmlFinal = $doc->saveHTML();

print_r(htmlspecialchars($htmlFinal));             
echo '<hr/><hr/><hr/>';

return $htmlFinal;
}

    Total execution time in seconds: 0.00026822090148926

Comments are closed.