Improved the webfinger handler for the #GoActivityPub reference implementation to support finding https:// resources alongside acct: double @ ones used by the larger fediverse.
In #FedBOX land, I have finally made progress in excising the OAuth2 routing to its own individual service.
Deploying has increased complexity, but it paves the path for replacing our spit and duct-tape user management to something more robust, like #ory#kratos.
Leftovers in FedBOX are the OAuth2 client CRUD operations, soon to be replaced with automated client creation.
I made some more progress on the application I plan to bundle as an admin tool with #FedBOX, the generic #ActivityPub server, and #Oni, the no frills single user instance.
It's called #Motely and it looks like this currently.
My attempt at replacing the current filtering API in #GoActivityPub storage backends with an improved static typed one is becoming a bit of a chore. The number of exposed functions has ballooned a lot because of the large surface of the ActivityPub objects compounded by the number of filtering criteria we want to support (which are not even that many): equality, fuzzy search, null value.
Very productive day in the land of #GoActivityPub. I managed to take the first steps for an improved filtering API I wanted between the ActivityPub processing package and the storage backends.
Speaking of #ActivityPub things, I realized that one of my previous design decisions for the custom storage functionality of #GoActivityPub: to include both OAuth2 and ActivityPub API interfaces is now biting me in the butt when I'm trying to increase parallelism on the sqlite storage implementation.
Currently there's a single sqlite db for the whole instance - so write locking happens a lot.
1/2
#Oni, the single user federated instance based on #GoActivityPub, serves HTML and #ActivityPub content from the same URLs based on what the clients request.
Firefox was caching the responses without taking into consideration the Content-Type, and was serving the JSON instead of HTML if refreshing a page. Only today I was reminded of the Vary header and fixed the problem. Yay! :D
I'm a little bit ashamed of myself that I've forgotten about it.
One of the small things I managed to do while sick this week was to add support for <link rel=alternate type="application/activity+json"> for all the pages where this makes sense.
So now the instance and the individual users are directly discoverable on Mastodon instances by entering the URL in the search box. (Accepting the follows is not functional at the moment... oops)
One of the small things I managed to do while sick this week was to add support for <link rel=alternate type="application/activity+json"> for all the pages where this makes sense.
So now the instance and the individual users are directly discoverable on Mastodon instances by entering the URL in the search box. (Accepting the follows is not functional at the moment... oops)
Stemming from a (not very constructive) discussion I had this morning about SPAM in the #fediverse, I started looking at an integration for #GoActivityPub projects with #SpamAssassin.
I realized we need a portmanteau for #activitypub filter similar to the existing milter (mail filter). :D Any thoughts?