Merge pull request #21 from twingly/remove-scheme

Twingly::URL::Utilities.remove_scheme

README.md

@@ -15,6 +15,8 @@ Twingly URL tools.
     * `Twingly::URL::Hasher.documentdb_hash(url)` - SHA256 unsigned long, native endian digest
     * `Twingly::URL::Hasher.autopingdb_hash(url)` - SHA256 64-bit signed, native endian digest
     * `Twingly::URL::Hasher.pingloggerdb_hash(url)` - SHA256 64-bit unsigned, native endian digest
+* `twingly/url/utilities` - Utilities to work with URLs
+    * `Twingly::URL::Utilities.remove_scheme(url)` - Removes scheme from HTTP/HTTPS URLs (`http://twingly.com` -> `//twingly.com`)
 
 ## Normalization example
 

lib/twingly/url/utilities.rb

@@ -0,0 +1,13 @@
+module Twingly
+  module URL
+    module Utilities
+      module_function
+
+      PROTOCOL_EXPRESSION = /^https?:/i
+
+      def remove_scheme(url)
+        url.sub(PROTOCOL_EXPRESSION, '')
+      end
+    end
+  end
+end

test/test_helper.rb

@@ -5,3 +5,4 @@
 require 'twingly/url'
 require 'twingly/url/hasher'
 require 'twingly/url/normalizer'
+require 'twingly/url/utilities'

test/unit/utilities_test.rb

@@ -0,0 +1,82 @@
+require 'test_helper'
+
+class TestUtilities < MiniTest::Unit::TestCase
+  context ".normalize" do
+    should "not remove scheme from non HTTP(S) URLs" do
+      url = 'gopher://www.duh.se/'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal 'gopher://www.duh.se/', result
+    end
+
+    should "remove scheme from mixed case HTTP URL" do
+      url = 'HttP://www.duh.se/'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//www.duh.se/', result
+    end
+
+    should "remove scheme from mixed case HTTPS URL" do
+      url = 'hTTpS://www.duh.se/'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//www.duh.se/', result
+    end
+
+    should "remove scheme from lowercase HTTP URL" do
+      url = 'http://www.duh.se/'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//www.duh.se/', result
+    end
+
+    should "remove scheme from lowercase HTTPS URL" do
+      url = 'https://www.duh.se/'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//www.duh.se/', result
+    end
+
+    should "remove scheme from uppercase HTTP URL" do
+      url = 'HTTP://WWW.DUH.SE/'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//WWW.DUH.SE/', result
+    end
+
+    should "remove scheme from uppercase HTTPS URL" do
+      url = 'HTTPS://WWW.DUH.SE/'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//WWW.DUH.SE/', result
+    end
+
+    should "remove scheme from URL with non ASCII characters" do
+      url = 'http://www.thecloset.gr/people/bloggers-pick-ιωάννα-τσιγαρίδα'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//www.thecloset.gr/people/bloggers-pick-ιωάννα-τσιγαρίδα', result
+    end
+
+    should "only remove scheme from HTTP URL" do
+      url = 'http://feedjira.herokuapp.com/?url=http://developer.twingly.com/feed.xml'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//feedjira.herokuapp.com/?url=http://developer.twingly.com/feed.xml', result
+    end
+
+    should "only remove scheme from HTTPS URL" do
+      url = 'https://feedjira.herokuapp.com/?url=https://signalvnoise.com/posts.rss'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal '//feedjira.herokuapp.com/?url=https://signalvnoise.com/posts.rss', result
+    end
+
+    should "not remove scheme from non HTTP(S) URLs with parameter" do
+      url = 'ftp://ftp.example.com/?url=https://www.example.com/'
+
+      result = Twingly::URL::Utilities.remove_scheme(url)
+      assert_equal 'ftp://ftp.example.com/?url=https://www.example.com/', result
+    end
+  end
+end