Open AramZS opened 8 years ago
Consider as part of implementing PF blocks
I've spent some time looking at this. I think it's too hard and too niche to build right now. Here's a summary of my thoughts for future review:
pf_feed
items, we store the feed URL. This might look like 'http://blog.sucuri.net/feed'
. We also collect the name of the site, which we get from the feed headers. This might look like 'Securi Blog'
. IMO this is not enough to build a meaningful blogroll, since clicking the links would take you to the feed URL. Some modern browsers don't have native handlers for RSS (Firefox, for example, forces a download, while Chrome shows you the file as raw text). This is bound to be confusing for most users. So a minimal version of a Blogroll probably needs to have the "publication" URL. This might be something like http://blog.securi.net
. The problem is that many/most feeds don't make this information available. We could use some heuristics to figure it out: feed URLs that look like WP can be parsed in a certain way, and we may be able to guess that the parent directory of a feed is the proper "publication" URL (eg https://example.com/foo/bar/rss.xml
becomes https://example.com/foo/bar/
). But this is subject to error, and will probably need some sort of override mechanism for manually entering the URL when editing the Subscribed Feed in the Dashboard.- Securi Blog [Site] [Feed]
- Washington Post [Site] [Feed]
So, in short, it's a pretty large amount of work to build something that is even minimally viable for a specific use case. Let's explore this in the future when we have a partner who can articulate such a use case.
Ability to show Blogroll, show by folder, show with alerts as their own tab.