Turn HTML named and numeric character references into Unicode code points.
Named, decimal, and hexadecimal character references are recognized by the HTML tokenizer and converted to Unicode code points, with browser-defined recovery for malformed or ambiguous references. The conversion is text decoding, not sanitization of the resulting content.
<div class="main">Tom & Jerry</div>
<div class="main">Tom & Jerry</div>
Scraped text, CMS previews, and browser-compatibility fixtures decode entities.