postrank-uri 1.0.5 → 1.0.6

data/lib/postrank-uri.rb

@@ -132,16 +132,32 @@ module PostRank
 
     def c18n(uri)
       u = parse(uri)
+      u = embedded(u)
 
       if q = u.query_values(:notation => :flat_array)
         q.delete_if { |k,v| C18N[:global].include?(k) }
         q.delete_if { |k,v| C18N[:hosts].find {|r,p| u.host =~ r && p.include?(k) } }
       end
-
       u.query_values = q
+
+      if u.host == 'twitter.com' && u.fragment.match(/^!(.*)/)
+        u.fragment = nil
+        u.path = $1
+      end
+
       u
     end
 
+    def embedded(uri)
+      if uri.host == 'news.google.com' && uri.path == '/news/url' \
+         || uri.host == 'xfruits.com'
+
+        embedded = uri.query_values['url']
+        uri = clean(embedded, false) if embedded
+      end
+      uri
+    end
+
     def parse(uri)
       return uri if uri.is_a? Addressable::URI
 

data/lib/postrank-uri/version.rb

@@ -1,5 +1,5 @@
 module PostRank
   module URI
-    VERSION = "1.0.5"
+    VERSION = "1.0.6"
   end
 end

data/spec/postrank-uri_spec.rb

@@ -112,7 +112,26 @@ describe PostRank::URI do
         c('igvita.com/?id=a&utm_source=a&awesm=b').should == 'http://igvita.com/?id=a'
         c('igvita.com/?id=a&sms_ss=a').should == 'http://igvita.com/?id=a'
       end
+    end
 
+    context "hashbang" do
+      it "should rewrite twitter links to crawlable versions" do
+        c('http://twitter.com/#!/igrigorik').should == 'http://twitter.com/igrigorik'
+        c('http://twitter.com/#!/a/statuses/1').should == 'http://twitter.com/a/statuses/1'
+        c('http://nontwitter.com/#!/a/statuses/1').should == 'http://nontwitter.com/#!/a/statuses/1'
+      end
+    end
+
+    context "embedded links" do
+      it "should extract embedded redirects from Google News" do
+        u = c('http://news.google.com/news/url?sa=t&fd=R&&url=http://www.ctv.ca/CTVNews/Politics/20110111/')
+        u.should == 'http://www.ctv.ca/CTVNews/Politics/20110111/'
+      end
+
+      it "should extract embedded redirects from xfruits.com" do
+        u = c('http://xfruits.com/MrGroar/?url=http%3A%2F%2Faap.lesroyaumes.com%2Fdepeches%2Fdepeche351820908.html')
+        u.should == 'http://aap.lesroyaumes.com/depeches/depeche351820908.html'
+      end
     end
   end
 
@@ -171,6 +190,10 @@ describe PostRank::URI do
       end
     end
 
+    it "should extract twitter links with hashbangs" do
+      e('test http://twitter.com/#!/igrigorik').should include('http://twitter.com/igrigorik')
+    end
+
     it "should handle a URL that comes after text without a space" do
       e("text:http://spn.tw/tfnLT").should include("http://spn.tw/tfnLT")
       e("text;http://spn.tw/tfnLT").should include("http://spn.tw/tfnLT")

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 1
   - 0
-  - 5
-  version: 1.0.5
+  - 6
+  version: 1.0.6
 platform: ruby
 authors: 
 - Ilya Grigorik
@@ -14,7 +14,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-02-02 00:00:00 -05:00
+date: 2011-02-22 00:00:00 -05:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency