rfeedfinder 0.9.1 → 0.9.2

data/History.txt

@@ -1,3 +1,7 @@
+== 0.9.2 2007-08-27
+
+* Verify HTTP::302 redirection, really useful for fotolog feeds
+
 == 0.9.1 2007-08-26
 
 * Can read Jumpcut webpage

data/lib/rfeedfinder.rb

@@ -71,6 +71,24 @@ module Rfeedfinder
       return newuris[0].strip
     end
   end
+  
+  def verifyRedirect(feedlist)
+    feedlist.each do |feed|
+      begin
+       response = Net::HTTP.get_response(URI.parse(feed))
+       #puts "Verify #{feed} - code: #{response.code}"
+       if response.code == "302"
+         newuri = response.body.match(/<a href=\"([^>]+)\">/)[1]
+         feedlist.delete(feed)
+         feedlist << newuri
+         feedlist.uniq!
+       end
+     rescue
+       # rescue net error
+     end
+    end
+    return feedlist
+  end
 
   def isFeedData?(data)
     # if no html tag and rss, rdf or feed tag, it's a feed
@@ -132,7 +150,11 @@ module Rfeedfinder
     end
 
     # is this already a feed?
-    return [fulluri] if isFeedData?(data)
+    if isFeedData?(data)
+      feedlist = [fulluri]
+      verifyRedirect(feedlist)
+      return feedlist
+    end
     
     #verify redirection
     newuri = tryBrokenRedirect(data)
@@ -219,6 +241,10 @@ module Rfeedfinder
     #_debuglog('still no luck, searching Syndic8')
     outfeeds << getFeedsFromSyndic8(uri) if querySyndic8 and outfeeds.empty?
     #outfeeds = list(set(outfeeds)) if hasattr(__builtins__, 'set') or __builtins__.has_key('set')
+    
+    # Verify redirection
+    verifyRedirect(outfeeds)
+    
     return outfeeds.flatten
   end
 

data/lib/rfeedfinder/version.rb

@@ -2,7 +2,7 @@ module Rfeedfinder #:nodoc:
   module VERSION #:nodoc:
     MAJOR = 0
     MINOR = 9
-    TINY  = 1
+    TINY  = 2
 
     STRING = [MAJOR, MINOR, TINY].join('.')
   end

data/test/test_rfeedfinder.rb

@@ -95,10 +95,22 @@ class TestRfeedfinder < Test::Unit::TestCase
   def test_fotolog
     feed_finder "http://www.fotolog.com/darth_fonsu/"
     feed_finder "http://www.fotolog.com/darth_fonsu/feed/main/rss20"
-    feed_finder "http://www1.fotolog.com/mad_lux", "http://www.fotolog.com/mad_lux/feed/main/rss20"
+    feed_finder "http://www1.fotolog.com/mad_lux", "http://www0.fotolog.com/mad_lux/feed/main/rss20"
     feed_finder "http://www1.fotolog.com/kel_06/"
   end
   
+  def test_google_video
+    feed_finder "http://video.google.com/videosearch?hl=en&safe=off&q=the+office"
+  end
+  
+  def test_blogsome
+    feed_finder "http://voxd.blogsome.com/"
+  end
+  
+  def test_bloger_hr
+    feed_finder "http://andreja666.bloger.hr/"
+  end
+  
   def test_not_fulluri_link
     # Meta link is only giving /feed/atom.xml
     feed_finder "http://blog.zvents.com/", "http://blog.zvents.com/feed/atom.xml"

data/website/index.html

@@ -33,7 +33,7 @@
     <h1>rfeedfinder</h1>
     <div id="version" class="clickable" onclick='document.location = "http://rubyforge.org/projects/rfeedfinder"; return false'>
       <p>Get Version</p>
-      <a href="http://rubyforge.org/projects/rfeedfinder" class="numbers">0.9.1</a>
+      <a href="http://rubyforge.org/projects/rfeedfinder" class="numbers">0.9.2</a>
     </div>
     <h2>What</h2>
 

metadata

@@ -3,8 +3,8 @@ rubygems_version: 0.9.4
 specification_version: 1
 name: rfeedfinder
 version: !ruby/object:Gem::Version 
-  version: 0.9.1
-date: 2007-08-25 00:00:00 +02:00
+  version: 0.9.2
+date: 2007-08-29 00:00:00 +02:00
 summary: rFeedFinder uses RSS autodiscovery, Atom autodiscovery, spidering, URL correction, and Web service queries -- whatever it takes -- to find the feed.
 require_paths: 
 - lib