I spent about five hours last night playing with my RSS Reader. I think I’m beginning to get the hang of some of this PHP stuff. Anyways, I’ve made some tweaks here and there to make it a bit more friendly.
The first thing I did was make it more error-friendly. If someone mistypes a feed URL, instead of seeing nonsense errors that have no value to the user, you get a simple error page that suggests that there was a problem with the feed, and to try a new feed URL.
The next thing I did was allow the user to skip the http:// protocol as long as there was a www. at the beginning of it. If the URL doesn’t have a www., then the http:// is required.
I also added support for the feed:// protocol. I did two things with this. Since the protocol is the same as http:// anyways (it’s designed to help feed readers subscribe a feed), I have the RSS Reader strip out the feed://, and replace it with http:// so that Magpie can parse it. The other thing I did was add an option to subscribe to a particular feed on that feed’s page (i.e. Zeldman). This should be fairly convenient.
Another thing that I’ve made an attempt to support is auto-detection of RSS feeds. Sites that have something along the lines of this:
<link rel="alternate" type="application/rss+xml"
title="RSS" href="/rss/skyzyx.xml" />
It seems to work for some sites, but not for others. I’m still working on why the inconsistency exists, but I plan to have it worked out as soon as I get around to it. Some examples of sites that auto-detect are: The Daily Report, StopDesign, and Simon Willison. Some examples that don’t work are Superfluous Banter, Whitespace, and my own Skyzyx.com.
I’ve modified the user-agent string. To find out what it will return on your sites, load a valid feed, then check the source. I’ve noted it in the comments. Anyways, I hope that this tool is useful to someone.
Speaking of, what do you think about online RSS/Atom readers? Are they better or worse than desktop applications? What are some pros or cons of each?
