Jessica

trying to interact with the API of my feed reader service and feeling like it was written by aliens

I want to find the subset of feed entries which are:

1. unread,
2. belong to a particular folder of feeds, and
3. older than 𝑥 date

this appears to require requesting a complete list of unread entry IDs, and then either the complete list of entries, or entries for each target feed, OR each individual entry, and then checking they match

entry objects do not have a read state flag, that is entirely the domain of the list of unread entry IDs

and I can see from a database structure perspective why you might do that but also oof ouch ow my application will need to store and compute SO much state

wellllll, I wondered how the feed reader's own web UI handles it and it turns out... it cheats, it just has a bespoke endpoint which returns baked HTML of the unread entries, no futzing with the list of unread IDs, boo

then I realised that the source is published on GitHub, and I had a peek at the source and... it's not doing anything special to achieve that

class UnreadEntriesController < ApplicationController
  def update
    # there is code in this method but it's not relevant
  end
end

def index
  @user = current_user
  render json: @user.unread_entries.pluck(:entry_id).compact.to_json
end