WordPress.com’s new editor was updated on November 16 and it broke redirecting for private and Jetpack-enabled blogs. Specifically, the fallback link to the old editor was removed, so the script could not scrape it and redirect to it.
Those of you who are using the script already and have automatic updates enabled (it’s probably enabled by default) should get an update to v1.4.0 soon. Otherwise, head over to Greasy Fork and install the script manually.
As a result of the update, redirection for Jetpack-enabled blogs whose site root and installation root differ fails again. Until someone figures out how to hijack the API proxy that the editor uses, this will stay broken.
Please report any issues you come across and I will try to fix them as soon as possible.
To go into a little technical detail about the changes in the script, the biggest change is in how private and Jetpack-enabled blogs are handled. Previously, the script would scrape the page for the fallback edit link. Since that link has been removed, the URL has to be built. The new URLs use the domain name as the blog ID instead of the numeric ID, so that’s used to build the editor link when available. This means that redirects for private blogs now have the potential to be faster than before. However, when the URL still contains a numeric ID, the script scrapes the page for the domain, specifically in the “Visit Site” box in the top-left, and uses the site address that it gets to build the link.
As for Jetpack-enabled blogs, public Jetpack-enabled blogs are now handled the same way as public WordPress.com blogs since scraping the page for the site link would produce the same link as an API call.
Feel free to dig into the code (excuse the mess) and ask questions or send pull requests. Although mostly outdated, the in-depth look at how v1.0.2 worked is still somewhat relevant as the basic idea is still the same. Between then and now, functionality for handling pages and private blogs have been added.