WeChat ID
aaronpk_tv
I think I'm in favor of merging this as is, even if it still returns invalid URLs, because it won't return anything more broken than the current version. I do think it should properly escape URLs in the parsed result though, since that's what would happen if you included a link in an <a> tag in a browser that included spaces. (The browser converts those to %20.
Just wanted to chime in here again to say that my implementation now also includes a rel=alternate link to the ActivityStreams JSON representation of the page.
e.g.
https://aaronparecki.com/2018/07/12/10/indieauth
<link rel="alternate" type="application/activity+json"
href="https://aaronparecki.com/2018/07/12/10/indieauth.as2" />
That brings this up to at least 3 implementations that support it, making it a good candidate to incorporate into the spec.
