Feedbag is a Ruby library for the auto-discovery of syndicated feeds (RSS/Atom).
>> require "rubygems"
=> true
>> require "feedbag"
=> true
>> Feedbag.find "http://damog.nl/blog"
=> ["http://damog.net/blog/index.rss", "http://damog.net/blog/tags/feed", "http://damog.net/blog/tags/rfeed"]
>> Feedbag.feed?("google.com")
=> false
>> Feedbag.feed?("http://planet.debian.org/rss20.xml")
=> true
$ gem install feedbag
Or just grab feedbag.rb and use it on your own project:
$ wget http://github.com/damog/feedbag/raw/master/lib/feedbag.rb
So you want to know more about it.
OK, if the URL passed to the find method is a feed itself, that only feed URL will be returned.
>> Feedbag.find "github.com/damog.atom"
=> ["http://github.com/damog.atom"]
>>
Otherwise, it will always return LINK feeds first, A (anchor tags) feeds later. Between A feeds, the ones hosted on the same URL's host, will have larger priority:
>> Feedbag.find "http://ve.planetalinux.org"
=> ["http://feedproxy.google.com/PlanetaLinuxVenezuela", "http://rendergraf.wordpress.com/feed/", "http://rootweiller.wordpress.com/feed/", "http://skatox.com/blog/feed/", "http://kodegeek.com/atom.xml", "http://blog.0x29.com.ve/?feed=rss2&cat=8"]
>>
On your application you should only take the very first element of the array, most of the times:
>> Feedbag.find("planet.debian.org").first(3)
=> ["http://planet.debian.org/rss10.xml", "http://planet.debian.org/rss20.xml", "http://planet.debian.org/atom.xml"]
>>
(Try running that same example without the "first" method. That example's host is a blog aggregator, so it has hundreds of feed URLs:)
>> Feedbag.find("planet.debian.org").size
=> 104
>>
Feedbag will find them all, but it will return the most important ones on the first elements on the array returned.
>> Feedbag.find("cnn.com")
=> ["http://rss.cnn.com/rss/cnn_topstories.rss", "http://rss.cnn.com/rss/cnn_latest.rss", "http://rss.cnn.com/services/podcasting/robinmeade/rss.xml"]
>>
- Because it's cool.
- Because it only uses Nokogiri as dependency.
- Because it follows modern feed filename conventions (like those ones used by WordPress blogs, or Blogger, etc).
- Because it's a single file you can embed easily in your application.
- Because I liked Benjamin Trott's Feed::Find.
- Because I thought it would be good to have Feed::Find's functionality in Ruby.
- Because I thought it was going to be easy to maintain.
- Because I was going to use it on rFeed.
- And finally, because I didn't know rfeedfinder existed :-)
Please, report bugs to rt@support.axiombox.com or directly to the author.
git clone git://github.com/damog/feedbag.git
...patch, build, hack and make pull requests. I'll be glad.
David Moreno <david@axiombox.com>.
This is free software. See COPYING for more information.
Raquel, for making Axiombox and most of my dreams possible. Also, GitHub for making a nice code sharing service that doesn't suck.