unshorten 0.2.1 → 0.3.0

checksums.yaml

@@ -1,15 +1,7 @@
 ---
-!binary "U0hBMQ==":
-  metadata.gz: !binary |-
-    ZTA5ZWQ3MDQ1YWM5YjVkMWE4ZWM0MWRlNmIxNTAyM2QyYmE1Njg2Mw==
-  data.tar.gz: !binary |-
-    MTExOThkNzU4YzhlM2VmYWYzMmVlZjg4NGQyZTIyOTkxZWYzZDI2NQ==
-!binary "U0hBNTEy":
-  metadata.gz: !binary |-
-    ZjFkZDBiYmNjOGY0MDI3MDFlM2VlYmU1MDE5OTNjODk2YzUxNDBmMzJjMmZj
-    MWEzMDBhZmRjNzExM2ZkOGE2MTU2MTU0ZDUzYjJjNjgwYTkyM2Q0MjI3ZDVh
-    NGFiYzgzNGE3ZDAwNWYzZTQ5NDU1MzQyODk0YzAxNjYyM2EyNjU=
-  data.tar.gz: !binary |-
-    Y2FkZGY5NDU2NGI2ZDI4ZGMyOTVhYTJhNzIwNjU2MGM3MzFiODY3M2E3OWQ4
-    ODQ5YmVjZjNkMjM3MjMxNGFlZGQwZDk5Y2I3ZDdkNWU0MWZmM2QyZWVkODRk
-    MDNkYTIwNGE2OTdiNjY5Y2Q4OTNkYzQxNDkyMGI3NjJhY2Y4ODY=
+SHA1:
+  metadata.gz: e9b5c2d553a05779d8a45cec5d288152a2ee9840
+  data.tar.gz: 4a0c1dfe4bbfe38e0c276023d1f1d1a64a0e9465
+SHA512:
+  metadata.gz: 40481e8187e84d1d7921ae79b52836e9033813d0cebf6b27c6b1b7e331c28eb8ad3677e8b1ae03cee2402e180c28cc2c6d58cd16a315afbba08933280d908c04
+  data.tar.gz: e4f1bd8db26b1801c8ffc57f503a0882197633faf9b95733fd2ecd32b1bfaf6c38b05f338b0622fd6669b360b3cedc50728be96ebda1473e4c41da5539f8bb3f

data/lib/unshorten.rb

@@ -12,7 +12,8 @@ class Unshorten
   DEFAULT_OPTIONS = {
     :max_level => 10,
     :timeout => 2,
-    :short_hosts => Regexp.union('url', /^[^.]{1,6}\.[^.]{1,2}$/i),
+    :short_hosts => false,
+    :short_urls => /^(?:https?:)?\/*[^\/]*\/*[^\/]*$/,
     :use_cache => true,
     :add_missing_http => true
   }
@@ -29,8 +30,12 @@ class Unshorten
     # @option options [Integer] :timeout Timeout in seconds, for every request
     # @option options [Regexp]  :short_hosts Hosts that provides short url
     #                                        services, only send requests if
-    #                                        host matches this regexp. Set to false
-    #                                        to follow all redirects.
+    #                                        host matches this regexp. Set to 
+    #                                        nil to follow all redirects.
+    # @option options [Regexp]  :short_urls  URLs that looks like a short one.
+    #                                        Only send requests when the URL
+    #                                        match this regexp. Set to nil to
+    #                                        follow all redirects.
     # @option options [Boolean] :use_cache Use cached result if available
     # @option options [Boolean] :add_missing_http add 'http://' if missing
     # @see DEFAULT_OPTIONS
@@ -69,11 +74,12 @@ class Unshorten
 
       return @@cache[url] if options[:use_cache] and @@cache[url]
       return url if level >= options[:max_level]
+      return url if options[:short_urls] && ! (url =~ options[:short_urls])
 
       uri = URI.parse(url) rescue nil
 
       return url if uri.nil?
-      return url if options[:short_hosts] != false and not uri.host =~ options[:short_hosts]
+      return url if options[:short_hosts] && ! (uri.host =~ options[:short_hosts])
 
       http = Net::HTTP.new(uri.host, uri.port)
       http.open_timeout = options[:timeout]
@@ -84,7 +90,9 @@ class Unshorten
 
       if response.is_a? Net::HTTPRedirection and response['location'] then
         expire_cache if @@cache.size > CACHE_SIZE_LIMIT
-        @@cache[url] = follow(response['location'], options, level + 1)
+        location = response['location']
+        location = (uri + location).to_s if location
+        @@cache[url] = follow(location, options, level + 1)
       else
         url
       end

data/test/test_unshorten.rb

@@ -1,42 +1,53 @@
 # coding: utf-8
 
-require 'test/unit'
+require 'minitest/autorun'
 require 'unshorten'
 
-class UnshortenTest < Test::Unit::TestCase
-    ORIGINAL_URL = 'http://dir.yahoo.com/Reference/Libraries/Library_and_Information_Science/Metadata/URIs___Universal_Resource_Identifiers/URLs___Uniform_Resource_Locators/URL_Shortening/'
-    SHORTENED_URL = 'http://tinyurl.com/j'
+class UnshortenTest < MiniTest::Test
+  ORIGINAL_URL = 'http://dir.yahoo.com/Reference/Libraries/Library_and_Information_Science/Metadata/URIs___Universal_Resource_Identifiers/URLs___Uniform_Resource_Locators/URL_Shortening/'
+  SHORTENED_URL = 'http://tinyurl.com/j'
 
-    def test_unshorten_alias
-        assert_equal ORIGINAL_URL, Unshorten[SHORTENED_URL, :use_cache => false]
-    end
+  def test_unshorten_alias
+    assert_equal ORIGINAL_URL, Unshorten[SHORTENED_URL, :use_cache => false]
+  end
 
-    def test_illegal_urls
-        illegal_url = 'http://a汉字bc.那么/是非法的?url'
-        assert_equal illegal_url, Unshorten.unshorten(illegal_url, :use_cache => false)
-    end
+  def test_illegal_urls
+    illegal_url = 'http://a汉字bc.那么/是非法的?url'
+    assert_equal illegal_url, Unshorten.unshorten(illegal_url, :use_cache => false)
+  end
 
-    def test_option_max_level
-        assert_equal SHORTENED_URL, Unshorten.unshorten(SHORTENED_URL, :max_level => 0, :use_cache => false)
-    end
+  def test_option_max_level
+    assert_equal SHORTENED_URL, Unshorten.unshorten(SHORTENED_URL, :max_level => 0, :use_cache => false)
+  end
 
-    def test_option_short_hosts
-        assert_equal SHORTENED_URL, Unshorten.unshorten(SHORTENED_URL, :short_hosts => /jmp/, :use_cache => false)
-    end
+  def test_option_short_hosts
+    assert_equal SHORTENED_URL, Unshorten.unshorten(SHORTENED_URL, :short_hosts => /jmp/, :use_cache => false)
+  end
 
-    HTTPS_LINK_1_FULL = 'https://github.com/aaronpk/unshorten'
-    HTTPS_LINK_1_SHORT = 'https://t.co/5204FqAr'
+  def test_option_short_urls
+    assert_equal SHORTENED_URL, Unshorten.unshorten(SHORTENED_URL, :short_urls => /DONT_MATCH/, :use_cache => false)
+    dummy_url = 'http://a.com/b/c'
+    assert_equal dummy_url, Unshorten.unshorten(dummy_url, :use_cache => false)
+  end
 
-    def test_https_link
-        assert_equal HTTPS_LINK_1_FULL, Unshorten.unshorten(HTTPS_LINK_1_SHORT, :short_hosts => /./, :use_cache => false)
-    end
+  HTTPS_LINK_1_FULL = 'https://github.com/aaronpk/unshorten'
+  HTTPS_LINK_1_SHORT = 'https://t.co/5204FqAr'
 
-    def test_follow_all_redirects
-        assert_equal HTTPS_LINK_1_FULL, Unshorten.unshorten(HTTPS_LINK_1_SHORT, :short_hosts => false, :use_cache => false)
-    end
+  def test_https_link
+    assert_equal HTTPS_LINK_1_FULL, Unshorten.unshorten(HTTPS_LINK_1_SHORT, :short_hosts => /./, :use_cache => false)
+  end
+
+  def test_follow_all_redirects
+    assert_equal HTTPS_LINK_1_FULL, Unshorten.unshorten(HTTPS_LINK_1_SHORT, :short_hosts => false, :use_cache => false)
+  end
+
+  def test_only_tco_links
+    assert_equal "http://aaron.pk/649", Unshorten.unshorten(HTTPS_LINK_1_SHORT, :short_hosts => /t\.co/, :use_cache => false)
+  end
+
+  def test_relative_url
+    assert_equal 'http://waterpigs.co.uk/articles/getting-started-with-microformats2/', Unshorten.unshorten('http://bit.ly/1l1Dz6K', :short_urls => nil, :use_cache => false)
+  end
 
-    def test_only_tco_links
-        assert_equal "http://aaron.pk/649", Unshorten.unshorten(HTTPS_LINK_1_SHORT, :short_hosts => /t\.co/, :use_cache => false)
-    end
 
 end

metadata

@@ -1,15 +1,43 @@
 --- !ruby/object:Gem::Specification
 name: unshorten
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.3.0
 platform: ruby
 authors:
-- Wu Jun
+- Jun Wu
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-04-05 00:00:00.000000000 Z
-dependencies: []
+date: 2014-03-09 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: minitest
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 description: Get original URLs from shortened ones
 email: quark@lihdd.net
 executables:
@@ -17,9 +45,9 @@ executables:
 extensions: []
 extra_rdoc_files: []
 files:
+- bin/unshorten
 - lib/unshorten.rb
 - test/test_unshorten.rb
-- bin/unshorten
 homepage: https://github.com/quark-zju/unshorten
 licenses: []
 metadata: {}
@@ -29,20 +57,19 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ! '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ! '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.3
+rubygems_version: 2.2.0
 signing_key: 
 specification_version: 4
 summary: Unshorten URLs
 test_files:
 - test/test_unshorten.rb
-has_rdoc: