Quite a few rules were updated/enhanced, some added, some dead hostings removed
ShadowDOM support added for sites built with Web Components e.g. Polymer
Ancient browsers aren't supported because the code was refactored to the common JS norms and ES2015+ syntax
Click MPIV: Configure in your userscript manager popup menu.
The internal status updates are not exposed by default on the <html> node because doing so slows down complex sites due to recalculation of the entire page layout. Instead only the hovered node (as reported by the matching rule) receives status updates on its mpiv-status attribute (it's not the class nor data- attribute to avoid confusing sites with unknown stuff being present in these standard places). If you were using the global status feature to customize CSS of those statuses, you'll need to enable it manually in the MPIV's config dialog.
New rule property "u" (a single string or an array of strings) that performs a very fast plain-string check. Only when it succeeds, the slow regexp "r" is checked. Special symbols may be specified in "u" property to increase the reliability of matching: ||, |, ^ - same syntax as in AdBlock filters, see the source code of the script for usage examples.
||foo.bar/path, here || means "domain or subdomain" so the pattern matches domains like foo.bar or subdomain.foo.bar and doesn't match unrelated domains partially like for example foofoo.bar
|foo matches things that start with foo (the entire URL is checked so that means http at least, usually)
^ is a URL part separator (like / or ? or :) but not a letter/number, neither any of %._-. Additionally, when used at the end like foo^ it also matches when the source ends with foo