pretty_proxy 2.0.1 → 3.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 5c71de1a7325089cea6ee46ebd45882e65119619
-  data.tar.gz: 79bff31288cc6c7723e338e59bf010319b8e2d42
+  metadata.gz: 9ba185c1e9ef80173adf3ed23ad9b0ec056e305e
+  data.tar.gz: 4af0397bcdcaa6cd5baa7a2ea664e1ba1e734449
 SHA512:
-  metadata.gz: be855fc2ac44bcb2e9cea9b7714288d5fc333e66863bd63deaa5133790d1c65e10bb5e602f53f7ae2c28314a92c9c785ca3c4d77666bf8fe4844ec19884044c6
-  data.tar.gz: e19ea8b24a83f2d20a87d25776da3d77997bab40f0ceb955995888e68f0243f50a25b90d4893096c67c3497b40c3c67a3b5197de8a623731e5bb5d2b0a41d188
+  metadata.gz: a3f0e0c339617821d7122b1cfe8784e34f17b1f4c70e628c0f670061b71b8ba900cf32c7a202fe26b998b26b182312c50001810f36378b7456e73dec5db0ff74
+  data.tar.gz: da92cb376d0975d95faf9ab262b24405ebd2156da6d4a0170ab693e2d73247c7d7b2eafd1482005eb24d88b037e368966e2ef02cf3841334493c390a2f63a4a5

data/lib/pretty_proxy.rb

@@ -1,8 +1,9 @@
 require 'pathname'
-require 'uri'
 require 'nokogiri'
 require 'rack'
 require 'rack-proxy'
+require 'addressable/uri'
+require 'uri'
 
 # The PrettyProxy class aggregate and validate the configuration of a 
 # proxy based in simple pretty url oriented rewriting rules. It's too
@@ -47,9 +48,18 @@ require 'rack-proxy'
 # the proxy_path and is followed by a original_path.
 #
 # 'in(side)/out(side) the proxy control': The url have (or not) the path
-# starting with a original_path and the scheme, port and host are the
+# starting with a original_path, and the scheme, port and host are the
 # same of the original_domain.
 #
+# CHANGELOG:
+#   3.0.0
+#     * return a String for unproxify_url (and not more a URI)
+#        because this is a change in the API (and can break code) the major
+#        version is now 3, if you don't use this method you can safely upgrade
+#     * depends in addressable gem
+#     * handles correctly the URIs without scheme (but with host)
+#       like '//duckduckgo.com/' (spec added for that)
+#
 # @author Henrique Becker
 class PrettyProxy < Rack::Proxy
   # The supertype of any exceptions explicitly raised by the methods
@@ -83,7 +93,7 @@ class PrettyProxy < Rack::Proxy
     Utils.validate_original_domain_and_paths(original_domain, original_paths)
 
     @proxy_path = proxy_path.clone
-    @original_domain = URI(original_domain.clone)
+    @original_domain = Addressable::URI.parse(original_domain.clone)
     @original_paths = Set.new 
     if original_paths.respond_to? :each
       original_paths.each { | value | @original_paths << value.clone }
@@ -121,29 +131,21 @@ class PrettyProxy < Rack::Proxy
   # Take a proxy url and return the original URL behind the proxy. Preserve the
   # query and fragment, if any. For the rewrite of a request @see rewrite_env.
   # @param url [String, URI::HTTP, URI::HTTPS] A URL.
-  # @return [URI::HTTP, URI::HTTPS] A URI object.
+  # @return [String] The unproxified URI in a string.
   # @raise PrettyProxy::ProxyError
   def unproxify_url(url)
-    url = URI(url.clone)
-    unless url.path.start_with?(@proxy_path)
-      fail ProxyError, "url path has to be prefixed by proxy_path (#{@proxy_path})"
-    end
-    url.path = url.path.slice((proxy_path.size-1)..-1)
-    unless original_paths.any? { | path | url.path.start_with? path }
-      fail ProxyError, "the proxy only responds to paths in the original_paths (#{@original_paths})"
-    end
-    if url.host == original_domain.host && url.path.start_with?(@proxy_path)
-      fail ProxyError, 'this is a request for the proxy for a proxy page (recursive request)'
+    url = Addressable::URI.parse(url.clone)
+    
+    unless valid_path_for_proxy? url.path
+      fail ProxyError, "'#{url.to_s}' isn't inside the proxy control, it can't be unproxified"
     end
-    url.host = original_domain.host
-    url.scheme = original_domain.scheme
-    url.port = original_domain.port
 
-    url
-  rescue URI::InvalidURIError
+    url.site = @original_domain.site
+    url.path = url.path.slice((@proxy_path.size-1)..-1)
+
+    url.to_s
+  rescue Addressable::URI::InvalidURIError
     raise ArgumentError, "the url argument isn't a valid uri"
-  rescue URI::Error => e
-    raise ProxyError, "an unexpected URI exception has been thrown, the message is '#{e.message}'"
   end
 
   # Take a hyperlink and the url of the proxy page (not the original page)
@@ -158,34 +160,38 @@ class PrettyProxy < Rack::Proxy
   # @return [String] A relative path or an url.
   # @raise PrettyProxy::ProxyError
   def proxify_hyperlink(hyperlink, proxy_page_url)
-    hyperlink = URI(hyperlink.clone)
-    proxy_page_url = URI(proxy_page_url)
-    if hyperlink.scheme
-      if inside_proxy_control? hyperlink
-        unless point_to_a_proxy_page?(hyperlink, proxy_page_url)
-          hyperlink = proxify_uri(hyperlink, proxy_page_url)
-        end
-      end
-    else
-      # recreate the original site url from the relative/absolute path
-      absolute_link = unproxify_url proxy_page_url
-      absolute_link.path = Pathname.new(absolute_link.path).join(hyperlink.path).to_s
-      if inside_proxy_control? absolute_link
+    hyperlink = Addressable::URI.parse(hyperlink.clone)
+    proxy_page_url = Addressable::URI.parse(proxy_page_url)
+
+    # this is URI relative ('//duckduckgo.com', '/path', '../path')
+    if hyperlink.relative?
+      absolute_hyperlink = Addressable::URI.parse(unproxify_url(proxy_page_url))
+                                           .join(hyperlink)
+      if inside_proxy_control? absolute_hyperlink
+        # this is path relative ('../path', 'path', but not '//duckduckgo.com' or '/path')
         if Pathname.new(hyperlink.path).relative?
-          if same_domain_as_original?(proxy_page_url) &&
-               valid_path_for_proxy?(absolute_link.path)
+          if point_to_a_proxy_page?(absolute_hyperlink, proxy_page_url)
             # in the case of a relative path in the original page who points
             # to a proxy page, and the proxy page is inside the proxy control
-            # we have to use the absolute_link or the page will be double proxified
-            # example: ../proxy/content in http://example.com/proxy/content, with
-            # original_path as '/' is http://example.com/proxy/proxy/content
-            hyperlink = absolute_link
+            # we have to use the absolute_hyperlink or the page will be double
+            # proxified. Example: ../proxy/content in http://example.com/proxy/content,
+            # with original_path as '/' is http://example.com/proxy/proxy/content
+            hyperlink = absolute_hyperlink
           end
-        else # is absolute
-          hyperlink.path = @proxy_path + hyperlink.path[1..-1]
+        else
+          hyperlink.path = @proxy_path[0..-2] + absolute_hyperlink.path
+          hyperlink.host = proxy_page_url.host if hyperlink.host
+          hyperlink.port = proxy_page_url.port if hyperlink.port
         end
       else
-        hyperlink = absolute_link
+        hyperlink = absolute_hyperlink
+      end
+    else # the hyperlink is absolute
+      if inside_proxy_control? hyperlink
+        # if points to the proxy itself we don't double-proxify
+        unless point_to_a_proxy_page?(hyperlink, proxy_page_url)
+          hyperlink = proxify_uri(hyperlink, proxy_page_url)
+        end
       end
     end
 
@@ -238,7 +244,8 @@ class PrettyProxy < Rack::Proxy
   def rewrite_env(env)
     env = env.clone
     url_requested_to_proxy = Rack::Request.new(env).url
-    unproxified_url = unproxify_url(url_requested_to_proxy)
+    # Using URI, and not Addressable::URI because the port value is incorrect in the last
+    unproxified_url = URI(unproxify_url(url_requested_to_proxy))
 
     if env['HTTP_HOST']
       env['HTTP_HOST'] = unproxified_url.host
@@ -402,12 +409,10 @@ class PrettyProxy < Rack::Proxy
   end
 
   # api private Don't use this method.
-  def proxify_uri(uri, proxy_domain)
-    uri = uri.clone
+  def proxify_uri(absolute_uri, proxy_site)
+    uri = absolute_uri.clone
 
-    uri.scheme = proxy_domain.scheme
-    uri.host = proxy_domain.host
-    uri.port = proxy_domain.port
+    uri.site = proxy_site.site
     uri.path = @proxy_path + uri.path[1..-1]
 
     uri
@@ -416,6 +421,7 @@ class PrettyProxy < Rack::Proxy
 
   # api private Don't use the methods of this class. They are for internal use only.
   module Utils
+    # TODO: check if isn't the case of change this for 'u1.site == u2.site'
     def self.same_domain?(u1, u2)
       u1.scheme == u2.scheme &&
         u1.host == u2.host &&
@@ -437,23 +443,23 @@ class PrettyProxy < Rack::Proxy
     def self.validate_original_domain_and_paths(original_domain, original_paths)
       fail ConfigError, 'original_paths is empty' if original_paths.empty?
 
-      original_domain = URI(original_domain) # can raise URI:Error's
+      original_domain = Addressable::URI.parse(original_domain)
       fail ConfigError, 'the original_domain has to have no query or fragment' if original_domain.query || original_domain.fragment
 
-      # can raise URI:Error's
+      # can raise errors
       test_uri = original_domain.clone
       if original_paths.respond_to?(:each)
-        original_paths.each { | path | test_uri.path = path }
+        original_paths.each do | path |
+          fail ConfigError, "original_paths value '#{path}' don't start with a '/'" unless path.start_with? '/'
+          test_uri.path = path
+        end
       else
+        fail ConfigError, "original_paths value '#{original_paths}' don't start with a '/'" unless original_paths.start_with? '/'
         test_uri.path = original_paths
       end
 
-    rescue URI::InvalidComponentError => e
-      raise ConfigError, "the original_paths contain a invalid path, message of the URI exception: '#{e.message}'"
-    rescue URI::InvalidURIError => e
-      raise ConfigError, "the original_domain isn't a valid URI, message of the URI exception: '#{e.message}'"
-    rescue URI::Error => e
-      raise ConfigError, "a unexpected URI::Error exception was raised, message of the exception: '#{e.message}'"
+    rescue Addressable::URI::InvalidURIError => e
+      raise ConfigError, "the message of the URI exception: '#{e.message}'"
     end
   end
 

data/spec/pretty_proxy_spec.rb

@@ -1,6 +1,7 @@
 require 'pretty_proxy'
 require 'equivalent-xml' # needed for be_equivalent_to xml rspec matcher
 require 'zlib'
+require 'uri'
 
 shared_examples 'an reader method who encapsulate a mutable variable' do
   context 'when the return is changed' do
@@ -94,7 +95,7 @@ describe PrettyProxy do
     end
 
     context 'when the original_domain is invalid' do
-      it { expect {new.call(right_args[0], 'http://myoriginalsite.com/%%%/', right_args[2])}.to raise_error(PrettyProxy::ConfigError)}
+      it { expect {new.call(right_args[0], 'http://<invalid>.com', right_args[2])}.to raise_error(PrettyProxy::ConfigError)}
     end
 
     context 'when the original_domain has a query' do
@@ -127,7 +128,7 @@ describe PrettyProxy do
   
   # NOTE: excessive metaprogramming? only save 3~6 lines
   [ [ :proxy_path=, "when proxy_path doesn't end with a '/'", '/proxy'],
-    [ :original_domain=, 'when the original_domain is invalid', 'http://myoriginalsite.com/%%%/'],
+    [ :original_domain=, 'when the original_domain is invalid', 'http://<invalid>.com'],
     [ :original_paths=, "when the original_paths don't begin with a '/'", 'content']
   ].each do | error_case |
     writter, context_desc, invalid_input = *error_case
@@ -144,31 +145,31 @@ describe PrettyProxy do
 
     context 'when the original_path has no trailing slash' do
       it 'allow no trailing slash in the url' do
-        expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1')).to eq URI('http://site.net/p1')
+        expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1')).to eq 'http://site.net/p1'
       end
       it 'allow trailing slash in the url' do
-        expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1/')).to eq URI('http://site.net/p1/')
+        expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1/')).to eq 'http://site.net/p1/'
       end
     end
     context 'when the original_path has a trailing slash' do
       it 'allow trailing slash in the url' do
-        expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p2/p2_2/')).to eq URI('http://site.net/p2/p2_2/')
+        expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p2/p2_2/')).to eq 'http://site.net/p2/p2_2/'
       end
       it "don't allow no trailing slash" do
         expect { pp.unproxify_url('http://myproxy.net/proxys/sitez/p2/p2_2') }.to raise_error(PrettyProxy::ProxyError)
       end
     end
     it 'allow subdirectories inside that path' do
-      expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1/a/b/c/')).to eq URI('http://site.net/p1/a/b/c/')
+      expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1/a/b/c/')).to eq 'http://site.net/p1/a/b/c/'
     end
     it 'preserve querys in the url' do
-      expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1/?q=error&l=pt')).to eq URI('http://site.net/p1/?q=error&l=pt')
+      expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1/?q=error&l=pt')).to eq 'http://site.net/p1/?q=error&l=pt'
     end
     it 'preserve fragments in the url' do
-      expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1/#id')).to eq URI('http://site.net/p1/#id')
+      expect(pp.unproxify_url('http://myproxy.net/proxys/sitez/p1/#id')).to eq 'http://site.net/p1/#id'
     end
     it 'change the port to the original' do
-      expect(pp.unproxify_url('http://myproxy.net:9292/proxys/sitez/p1/#id').port).to eq 80
+      expect(URI(pp.unproxify_url('http://myproxy.net:9292/proxys/sitez/p1/#id')).port).to eq 80
     end
 
     context 'when the url redirect to the own proxy' do
@@ -193,6 +194,9 @@ describe PrettyProxy do
     it "proxify hyperlinks with scheme to inside the proxy control" do
       expect(pp.proxify_hyperlink('http://site.net/p2/p2_2/', 'http://theproxy.net/proxy/p1')).to eq 'http://theproxy.net/proxy/p2/p2_2/'
     end
+    it "proxify hyperlinks without scheme to inside the proxy control" do
+      expect(pp.proxify_hyperlink('//site.net/p2/p2_2/', 'http://theproxy.net/proxy/p1')).to eq '//theproxy.net/proxy/p2/p2_2/'
+    end
     it "don't change hyperlinks with scheme to ouside the proxy control" do
       expect(pp.proxify_hyperlink('http://othersite.net', 'http://theproxy.net/proxy/p1')).to eq 'http://othersite.net'
     end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pretty_proxy
 version: !ruby/object:Gem::Version
-  version: 2.0.1
+  version: 3.0.0
 platform: ruby
 authors:
 - Henrique Becker
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-05-21 00:00:00.000000000 Z
+date: 2013-05-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
@@ -52,6 +52,20 @@ dependencies:
     - - ~>
       - !ruby/object:Gem::Version
         version: '0.3'
+- !ruby/object:Gem::Dependency
+  name: addressable
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '2.3'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '2.3'
 - !ruby/object:Gem::Dependency
   name: equivalent-xml
   requirement: !ruby/object:Gem::Requirement