Here is a little function that truncates a DomNode to a specified number of text characters. I use it to generate HTML excerpts for my blog entries.
<?php
function makehtmlexcerpt(DomNode $html, $excerptlength)
{
$remove = 0;
$htmllength = strlen(html_entity_decode($html->textContent, ENT_QUOTES, 'UTF-8'));
$truncate = $htmllength - $excerptlength;
if($htmllength > $excerptlength)
{
if($html->hasChildNodes())
{
$children = $html->childNodes;
for($counter = 0; $counter < $children->length; $counter ++)
{
$child = $children->item($children->length - ($counter + 1));
$childlength = strlen(html_entity_decode($child->textContent, ENT_QUOTES, 'UTF-8'));
if($childlength <= $truncate)
{
$remove ++;
$truncate = $truncate - $childlength;
}
else
{
$child = makehtmlexcerpt($child, $childlength - $truncate);
break;
}
}
if($remove != 0)
{
for($counter = 0; $counter < $remove; $counter ++)
{
$html->removeChild($html->lastChild);
}
}
}
else
{
if($html->nodeName == '#text')
{
$html->nodeValue = substr(html_entity_decode($html->nodeValue, ENT_QUOTES, 'UTF-8'), 0, $htmllength - $truncate);
}
}
}
return $html;
}
?>
