Time for an update!
The feed subscription dialog is now shown inside a modal, so you can add subscriptions no matter which view of HappyFeed you have currently open.
A big addition to my last update is the feed fetcher inside the subscription process. What this means is, that you can put a URL inside the input field and HappyFeed will fetch the RSS feed URL(s) of that website accordingly.
I find this pretty handy, since it's such a pain to search for RSS feeds on websites or search through their HTML source. If a website has more than one RSS feed (i.e. Wordpress has posts + comments feeds), HappyFeed will show you a list and let you choose which one you want to subscribe to.
I've build the drag & drop sidebar which i've mentioned in my last update. So you can now regroup your subscriptions. At this point it's only possible to place feeds in different groups. Later it should also be possible, to reorder subscriptions inside each group, but my priority on that is not so high right now.
Image Proxy for HTTPS Connections
If you want to run HappyFeed on your own Server with HTTPS/TLS enabled (which i highly recommend), you need to make sure all data is transfered over that connection. Since some RSS Feeds include images, and those images might come from Servers without HTTPS enabled, we have to let the HappyFeed Server proxy those images.
If we don't do that your HTTPS connection will break/leak information, since some of those Feed contents will come from unencrypted connections.
To handle those cases i've build a little image proxy that will handle those things for you. To make this possisble i use a pretty cool gem called html-pipeline. With html-pipeline i can rewrite the RSS content to let HappyFeed handle all image requests. I've build a html-pipeline filter that will fetch images from the websites server and proxy it to the HappyFeed user over HTTPS.
This ensures that images will always be served by your HappyFeed server and your HTTPS/TLS connection will not break. Even though this feature needs some further improvement (regarding caching and other file types) it works pretty good already