Adding support for public_suffix.

* New runtime dependency on `public_suffix` gem.
* Match the escaping rules in the `reg-name` rule.
  * Note: This implies underscores are no longer escaped in hostnames.

Author: sporkmonger

CHANGELOG.md

@@ -1,8 +1,10 @@
 # Addressable 2.5.0
-- Paths beginning with // and missing an authority are now considered invalid
-- URI validation now also takes place after setting a path
-- Handle backslashes in authority more like a browser for `heuristic_parse`
-- Unescaped backslashes in host now raise an `InvalidURIError`
+- add support for public suffixes and tld; first runtime dependency
+- hostname escaping should match RFC; underscores in hostnames no longer escaped
+- paths beginning with // and missing an authority are now considered invalid
+- validation now also takes place after setting a path
+- handle backslashes in authority more like a browser for `heuristic_parse`
+- unescaped backslashes in host now raise an `InvalidURIError`
 - `merge!`, `join!`, `omit!` and `normalize!` don't disable deferred validation
 - `heuristic_parse` now trims whitespace before parsing
 

addressable.gemspec

@@ -1,34 +1,37 @@
 # -*- encoding: utf-8 -*-
-# stub: addressable 2.4.0 ruby lib
+# stub: addressable 2.5.0 ruby lib
 
 Gem::Specification.new do |s|
   s.name = "addressable"
-  s.version = "2.4.0"
+  s.version = "2.5.0"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.require_paths = ["lib"]
   s.authors = ["Bob Aman"]
-  s.date = "2015-12-07"
+  s.date = "2016-11-04"
   s.description = "Addressable is a replacement for the URI implementation that is part of\nRuby's standard library. It more closely conforms to the relevant RFCs and\nadds support for IRIs and URI templates.\n"
   s.email = "bob@sporkmonger.com"
   s.extra_rdoc_files = ["README.md"]
-  s.files = ["CHANGELOG.md", "Gemfile", "LICENSE.txt", "README.md", "Rakefile", "addressable.gemspec", "data/unicode.data", "lib/addressable.rb", "lib/addressable/idna.rb", "lib/addressable/idna/native.rb", "lib/addressable/idna/pure.rb", "lib/addressable/template.rb", "lib/addressable/uri.rb", "lib/addressable/version.rb", "spec/addressable/idna_spec.rb", "spec/addressable/net_http_compat_spec.rb", "spec/addressable/rack_mount_compat_spec.rb", "spec/addressable/security_spec.rb", "spec/addressable/template_spec.rb", "spec/addressable/uri_spec.rb", "spec/spec_helper.rb", "tasks/clobber.rake", "tasks/gem.rake", "tasks/git.rake", "tasks/metrics.rake", "tasks/rspec.rake", "tasks/yard.rake"]
+  s.files = ["CHANGELOG.md", "Gemfile", "LICENSE.txt", "README.md", "Rakefile", "data/unicode.data", "lib/addressable", "lib/addressable.rb", "lib/addressable/idna", "lib/addressable/idna.rb", "lib/addressable/idna/native.rb", "lib/addressable/idna/pure.rb", "lib/addressable/template.rb", "lib/addressable/uri.rb", "lib/addressable/version.rb", "spec/addressable", "spec/addressable/idna_spec.rb", "spec/addressable/net_http_compat_spec.rb", "spec/addressable/rack_mount_compat_spec.rb", "spec/addressable/security_spec.rb", "spec/addressable/template_spec.rb", "spec/addressable/uri_spec.rb", "spec/spec_helper.rb", "tasks/clobber.rake", "tasks/gem.rake", "tasks/git.rake", "tasks/metrics.rake", "tasks/rspec.rake", "tasks/yard.rake"]
   s.homepage = "https://github.com/sporkmonger/addressable"
   s.licenses = ["Apache-2.0"]
   s.rdoc_options = ["--main", "README.md"]
   s.required_ruby_version = Gem::Requirement.new(">= 1.9.0")
-  s.rubygems_version = "2.5.0"
+  s.rubygems_version = "2.5.1"
   s.summary = "URI Implementation"
 
   if s.respond_to? :specification_version then
     s.specification_version = 4
 
     if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
+      s.add_runtime_dependency(%q<public_suffix>, [">= 2.0.3", "~> 2.0"])
       s.add_development_dependency(%q<bundler>, ["~> 1.0"])
     else
+      s.add_dependency(%q<public_suffix>, [">= 2.0.3", "~> 2.0"])
       s.add_dependency(%q<bundler>, ["~> 1.0"])
     end
   else
+    s.add_dependency(%q<public_suffix>, [">= 2.0.3", "~> 2.0"])
     s.add_dependency(%q<bundler>, ["~> 1.0"])
   end
 end

lib/addressable/uri.rb

@@ -18,6 +18,7 @@
 
 require "addressable/version"
 require "addressable/idna"
+require "public_suffix"
 
 ##
 # Addressable is a library for processing links and URIs.
@@ -44,7 +45,7 @@ module CharacterClasses
       UNRESERVED = ALPHA + DIGIT + "\\-\\.\\_\\~"
       PCHAR = UNRESERVED + SUB_DELIMS + "\\:\\@"
       SCHEME = ALPHA + DIGIT + "\\-\\+\\."
-      HOST = ALPHA + DIGIT + "\\-\\.\\[\\:\\]"
+      HOST = UNRESERVED + SUB_DELIMS + "\\[\\:\\]"
       AUTHORITY = PCHAR
       PATH = PCHAR + "\\/"
       QUERY = PCHAR + "\\/\\?"
@@ -1145,6 +1146,24 @@ def hostname=(new_hostname)
       self.host = v
     end
 
+    ##
+    # Returns the top-level domain for this host.
+    #
+    # @example
+    #   Addressable::URI.parse("www.example.co.uk").tld # => "co.uk"
+    def tld
+      PublicSuffix.parse(self.host, ignore_private: true).tld
+    end
+
+    ##
+    # Returns the public suffix domain for this host.
+    #
+    # @example
+    #   Addressable::URI.parse("www.example.co.uk").domain # => "example.co.uk"
+    def domain
+      PublicSuffix.domain(self.host, ignore_private: true)
+    end
+
     ##
     # The authority component for this URI.
     # Combines the user, password, host, and port components.

spec/addressable/uri_spec.rb

@@ -409,6 +409,10 @@ def to_s
     expect(@uri.normalized_host).to eq("example.com")
   end
 
+  it "returns 'com' for #tld" do
+    expect(@uri.tld).to eq("com")
+  end
+
   it "returns 'user:password@example.com:8080' for #authority" do
     expect(@uri.authority).to eq("user:password@example.com:8080")
   end
@@ -2344,6 +2348,80 @@ def to_s
   end
 end
 
+describe Addressable::URI, "when parsed from " +
+    "'HTTP://EXAMPLE.COM/'" do
+  before do
+    @uri = Addressable::URI.parse("HTTP://EXAMPLE.COM/")
+  end
+
+  it "should be equivalent to http://example.com" do
+    expect(@uri).to eq(Addressable::URI.parse("http://example.com"))
+  end
+
+  it "should correctly convert to a hash" do
+    expect(@uri.to_hash).to eq({
+      :scheme => "HTTP",
+      :user => nil,
+      :password => nil,
+      :host => "EXAMPLE.COM",
+      :port => nil,
+      :path => "/",
+      :query => nil,
+      :fragment => nil
+    })
+  end
+
+  it "should be identical to its duplicate" do
+    expect(@uri).to eq(@uri.dup)
+  end
+
+  it "should have an origin of 'http://example.com'" do
+    expect(@uri.origin).to eq('http://example.com')
+  end
+
+  it "should have a tld of 'com'" do
+    expect(@uri.tld).to eq('com')
+  end
+end
+
+describe Addressable::URI, "when parsed from " +
+    "'http://www.example.co.uk/'" do
+  before do
+    @uri = Addressable::URI.parse("http://www.example.co.uk/")
+  end
+
+  it "should have an origin of 'http://www.example.co.uk'" do
+    expect(@uri.origin).to eq('http://www.example.co.uk')
+  end
+
+  it "should have a tld of 'co.uk'" do
+    expect(@uri.tld).to eq('co.uk')
+  end
+
+  it "should have a domain of 'example.co.uk'" do
+    expect(@uri.domain).to eq('example.co.uk')
+  end
+end
+
+describe Addressable::URI, "when parsed from " +
+    "'http://sub_domain.blogspot.com/'" do
+  before do
+    @uri = Addressable::URI.parse("http://sub_domain.blogspot.com/")
+  end
+
+  it "should have an origin of 'http://sub_domain.blogspot.com'" do
+    expect(@uri.origin).to eq('http://sub_domain.blogspot.com')
+  end
+
+  it "should have a tld of 'com'" do
+    expect(@uri.tld).to eq('com')
+  end
+
+  it "should have a domain of 'blogspot.com'" do
+    expect(@uri.domain).to eq('blogspot.com')
+  end
+end
+
 describe Addressable::URI, "when parsed from " +
     "'http://example.com/~smith/'" do
   before do

tasks/gem.rake

@@ -20,6 +20,7 @@ namespace :gem do
 
     s.required_ruby_version = '>= 1.9.0'
 
+    s.add_runtime_dependency 'public_suffix', '~> 2.0', '>= 2.0.3'
     s.add_development_dependency 'bundler', '~> 1.0'
 
     s.require_path = "lib"
@@ -39,15 +40,8 @@ namespace :gem do
   desc "Generates .gemspec file"
   task :gemspec do
     spec_string = GEM_SPEC.to_ruby
-
-    begin
-      Thread.new { eval("$SAFE = 3\n#{spec_string}", binding) }.join
-    rescue
-      abort "unsafe gemspec: #{$!}"
-    else
-      File.open("#{GEM_SPEC.name}.gemspec", 'w') do |file|
-        file.write spec_string
-      end
+    File.open("#{GEM_SPEC.name}.gemspec", 'w') do |file|
+      file.write spec_string
     end
   end