TLDR: It can be used to capture all the paragraphs from some webnovel sites. *update: testing other sites, I've found it lacking.
This is intended to match a <p> tag and it's closing tag. It only matches and captures if there are no other html tags nested in the <p> tag.
There is room to adjust with the quantifier.
Condensed regex:
<p>(?![\n|<])(.{0,500}?)(?![^<][\s]|[\w]+[^>])</p>(?=\n|<div)