public_suffix 2.0.5 → 4.0.7

data/lib/public_suffix.rb

@@ -1,14 +1,16 @@
+# frozen_string_literal: true
+
 # = Public Suffix
 #
 # Domain name parser based on the Public Suffix List.
 #
-# Copyright (c) 2009-2017 Simone Carletti <weppos@weppos.net>
+# Copyright (c) 2009-2022 Simone Carletti <weppos@weppos.net>
 
-require "public_suffix/domain"
-require "public_suffix/version"
-require "public_suffix/errors"
-require "public_suffix/rule"
-require "public_suffix/list"
+require_relative "public_suffix/domain"
+require_relative "public_suffix/version"
+require_relative "public_suffix/errors"
+require_relative "public_suffix/rule"
+require_relative "public_suffix/list"
 
 # PublicSuffix is a Ruby domain name parser based on the Public Suffix List.
 #
@@ -20,46 +22,48 @@ require "public_suffix/list"
 # but was originally created to meet the needs of browser manufacturers.
 module PublicSuffix
 
-  DOT   = ".".freeze
-  BANG  = "!".freeze
-  STAR  = "*".freeze
+  DOT   = "."
+  BANG  = "!"
+  STAR  = "*"
 
   # Parses +name+ and returns the {PublicSuffix::Domain} instance.
   #
   # @example Parse a valid domain
   #   PublicSuffix.parse("google.com")
-  #   # => #<PublicSuffix::Domain ...>
+  #   # => #<PublicSuffix::Domain:0x007fec2e51e588 @sld="google", @tld="com", @trd=nil>
   #
   # @example Parse a valid subdomain
   #   PublicSuffix.parse("www.google.com")
-  #   # => #<PublicSuffix::Domain ...>
+  #   # => #<PublicSuffix::Domain:0x007fec276d4cf8 @sld="google", @tld="com", @trd="www">
   #
   # @example Parse a fully qualified domain
   #   PublicSuffix.parse("google.com.")
-  #   # => #<PublicSuffix::Domain ...>
+  #   # => #<PublicSuffix::Domain:0x007fec257caf38 @sld="google", @tld="com", @trd=nil>
   #
   # @example Parse a fully qualified domain (subdomain)
   #   PublicSuffix.parse("www.google.com.")
-  #   # => #<PublicSuffix::Domain ...>
+  #   # => #<PublicSuffix::Domain:0x007fec27b6bca8 @sld="google", @tld="com", @trd="www">
   #
-  # @example Parse an invalid domain
+  # @example Parse an invalid (unlisted) domain
   #   PublicSuffix.parse("x.yz")
-  #   # => PublicSuffix::DomainInvalid
+  #   # => #<PublicSuffix::Domain:0x007fec2f49bec0 @sld="x", @tld="yz", @trd=nil>
+  #
+  # @example Parse an invalid (unlisted) domain with strict checking (without applying the default * rule)
+  #   PublicSuffix.parse("x.yz", default_rule: nil)
+  #   # => PublicSuffix::DomainInvalid: `x.yz` is not a valid domain
   #
   # @example Parse an URL (not supported, only domains)
   #   PublicSuffix.parse("http://www.google.com")
-  #   # => PublicSuffix::DomainInvalid
+  #   # => PublicSuffix::DomainInvalid: http://www.google.com is not expected to contain a scheme
   #
   #
-  # @param  [String, #to_s] name The domain name or fully qualified domain name to parse.
-  # @param  [PublicSuffix::List] list The rule list to search, defaults to the default {PublicSuffix::List}
-  # @param  [Boolean] ignore_private
+  # @param  name [#to_s] The domain name or fully qualified domain name to parse.
+  # @param  list [PublicSuffix::List] The rule list to search, defaults to the default {PublicSuffix::List}
+  # @param  ignore_private [Boolean]
   # @return [PublicSuffix::Domain]
   #
-  # @raise [PublicSuffix::DomainInvalid]
-  #   If domain is not a valid domain.
-  # @raise [PublicSuffix::DomainNotAllowed]
-  #   If a rule for +domain+ is found, but the rule doesn't allow +domain+.
+  # @raise [PublicSuffix::DomainInvalid] If domain is not a valid domain.
+  # @raise [PublicSuffix::DomainNotAllowed] If a rule for +domain+ is found, but the rule doesn't allow +domain+.
   def self.parse(name, list: List.default, default_rule: list.default_rule, ignore_private: false)
     what = normalize(name)
     raise what if what.is_a?(DomainInvalid)
@@ -73,6 +77,7 @@ module PublicSuffix
     if rule.decompose(what).last.nil?
       raise DomainNotAllowed, "`#{what}` is not allowed according to Registry policy"
     end
+
     # rubocop:enable Style/IfUnlessModifier
 
     decompose(what, rule)
@@ -95,11 +100,11 @@ module PublicSuffix
   #   PublicSuffix.valid?("example.tldnotlisted")
   #   # => true
   #
-  # @example Validate a not-allowed domain
-  #   PublicSuffix.valid?("example.do")
-  #   # => false
-  #   PublicSuffix.valid?("www.example.do")
+  # @example Validate a not-listed domain with strict checking (without applying the default * rule)
+  #   PublicSuffix.valid?("example.tldnotlisted")
   #   # => true
+  #   PublicSuffix.valid?("example.tldnotlisted", default_rule: nil)
+  #   # => false
   #
   # @example Validate a fully qualified domain
   #   PublicSuffix.valid?("google.com.")
@@ -112,8 +117,8 @@ module PublicSuffix
   #   # => false
   #
   #
-  # @param  [String, #to_s] name The domain name or fully qualified domain name to validate.
-  # @param  [Boolean] ignore_private
+  # @param  name [#to_s] The domain name or fully qualified domain name to validate.
+  # @param  ignore_private [Boolean]
   # @return [Boolean]
   def self.valid?(name, list: List.default, default_rule: list.default_rule, ignore_private: false)
     what = normalize(name)
@@ -128,9 +133,9 @@ module PublicSuffix
   #
   # This method doesn't raise. Instead, it returns nil if the domain is not valid for whatever reason.
   #
-  # @param  [String, #to_s] name The domain name or fully qualified domain name to parse.
-  # @param  [PublicSuffix::List] list The rule list to search, defaults to the default {PublicSuffix::List}
-  # @param  [Boolean] ignore_private
+  # @param  name [#to_s] The domain name or fully qualified domain name to parse.
+  # @param  list [PublicSuffix::List] The rule list to search, defaults to the default {PublicSuffix::List}
+  # @param  ignore_private [Boolean]
   # @return [String]
   def self.domain(name, **options)
     parse(name, **options).domain
@@ -165,6 +170,7 @@ module PublicSuffix
     return DomainInvalid.new("Name is blank") if name.empty?
     return DomainInvalid.new("Name starts with a dot") if name.start_with?(DOT)
     return DomainInvalid.new("%s is not expected to contain a scheme" % name) if name.include?("://")
+
     name
   end
 

data/public_suffix.gemspec

@@ -12,14 +12,18 @@ Gem::Specification.new do |s|
   s.description = "PublicSuffix can parse and decompose a domain name into top level domain, domain and subdomains."
   s.licenses    = ["MIT"]
 
-  s.required_ruby_version = ">= 2.0"
+  s.metadata = {
+    "bug_tracker_uri" => "https://github.com/weppos/publicsuffix-ruby/issues",
+    "changelog_uri" => "https://github.com/weppos/publicsuffix-ruby/blob/master/CHANGELOG.md",
+    "documentation_uri" => "https://rubydoc.info/gems/#{s.name}/#{s.version}",
+    "homepage_uri" => s.homepage,
+    "source_code_uri" => "https://github.com/weppos/publicsuffix-ruby/tree/v#{s.version}",
+  }
+
+  s.required_ruby_version = ">= 2.3"
 
   s.require_paths    = ["lib"]
   s.files            = `git ls-files`.split("\n")
   s.test_files       = `git ls-files -- {test,spec,features}/*`.split("\n")
   s.extra_rdoc_files = %w( LICENSE.txt )
-
-  s.add_development_dependency "rake"
-  s.add_development_dependency "mocha"
-  s.add_development_dependency "yard"
 end

data/test/.empty

@@ -0,0 +1,2 @@
+# This is an empty file I use to force a non-empty commit when I only need to store notes
+..

data/test/acceptance_test.rb

@@ -1,16 +1,18 @@
+# frozen_string_literal: true
+
 require "test_helper"
 
 class AcceptanceTest < Minitest::Test
 
   VALID_CASES = [
-      ["example.com",             "example.com",        [nil, "example", "com"]],
-      ["foo.example.com",         "example.com",        ["foo", "example", "com"]],
+    ["example.com",             "example.com", [nil, "example", "com"]],
+    ["foo.example.com",         "example.com",        ["foo", "example", "com"]],
 
-      ["verybritish.co.uk",       "verybritish.co.uk",  [nil, "verybritish", "co.uk"]],
-      ["foo.verybritish.co.uk",   "verybritish.co.uk",  ["foo", "verybritish", "co.uk"]],
+    ["verybritish.co.uk",       "verybritish.co.uk",  [nil, "verybritish", "co.uk"]],
+    ["foo.verybritish.co.uk",   "verybritish.co.uk",  ["foo", "verybritish", "co.uk"]],
 
-      ["parliament.uk",           "parliament.uk",      [nil, "parliament", "uk"]],
-      ["foo.parliament.uk",       "parliament.uk",      ["foo", "parliament", "uk"]],
+    ["parliament.uk",           "parliament.uk",      [nil, "parliament", "uk"]],
+    ["foo.parliament.uk",       "parliament.uk",      ["foo", "parliament", "uk"]],
   ].freeze
 
   def test_valid
@@ -19,7 +21,11 @@ class AcceptanceTest < Minitest::Test
       trd, sld, tld = results
       assert_equal tld, parsed.tld, "Invalid tld for `#{name}`"
       assert_equal sld, parsed.sld, "Invalid sld for `#{name}`"
-      assert_equal trd, parsed.trd, "Invalid trd for `#{name}`"
+      if trd.nil?
+        assert_nil parsed.trd, "Invalid trd for `#{name}`"
+      else
+        assert_equal trd, parsed.trd, "Invalid trd for `#{name}`"
+      end
 
       assert_equal domain, PublicSuffix.domain(input)
       assert PublicSuffix.valid?(input)
@@ -28,10 +34,10 @@ class AcceptanceTest < Minitest::Test
 
 
   INVALID_CASES = [
-      ["nic.ke",                  PublicSuffix::DomainNotAllowed],
-      [nil,                       PublicSuffix::DomainInvalid],
-      ["",                        PublicSuffix::DomainInvalid],
-      ["  ",                      PublicSuffix::DomainInvalid],
+    ["nic.bd", PublicSuffix::DomainNotAllowed],
+    [nil,                       PublicSuffix::DomainInvalid],
+    ["",                        PublicSuffix::DomainInvalid],
+    ["  ",                      PublicSuffix::DomainInvalid],
   ].freeze
 
   def test_invalid
@@ -43,16 +49,16 @@ class AcceptanceTest < Minitest::Test
 
 
   REJECTED_CASES = [
-      ["www. .com",           true],
-      ["foo.co..uk",          true],
-      ["goo,gle.com",         true],
-      ["-google.com",         true],
-      ["google-.com",         true],
-
-      # This case was covered in GH-15.
-      # I decided to cover this case because it's not easily reproducible with URI.parse
-      # and can lead to several false positives.
-      ["http://google.com",   false],
+    ["www. .com", true],
+    ["foo.co..uk",          true],
+    ["goo,gle.com",         true],
+    ["-google.com",         true],
+    ["google-.com",         true],
+
+    # This case was covered in GH-15.
+    # I decided to cover this case because it's not easily reproducible with URI.parse
+    # and can lead to several false positives.
+    ["http://google.com",   false],
   ].freeze
 
   def test_rejected
@@ -66,9 +72,9 @@ class AcceptanceTest < Minitest::Test
 
 
   CASE_CASES = [
-      ["Www.google.com", %w( www google com )],
-      ["www.Google.com", %w( www google com )],
-      ["www.google.Com", %w( www google com )],
+    ["Www.google.com", %w( www google com )],
+    ["www.Google.com", %w( www google com )],
+    ["www.google.Com", %w( www google com )],
   ].freeze
 
   def test_ignore_case
@@ -84,35 +90,41 @@ class AcceptanceTest < Minitest::Test
 
 
   INCLUDE_PRIVATE_CASES = [
-      ["blogspot.com", true, "blogspot.com"],
-      ["blogspot.com", false, nil],
-      ["subdomain.blogspot.com", true, "blogspot.com"],
-      ["subdomain.blogspot.com", false, "subdomain.blogspot.com"],
+    ["blogspot.com", true, "blogspot.com"],
+    ["blogspot.com", false, nil],
+    ["subdomain.blogspot.com", true, "blogspot.com"],
+    ["subdomain.blogspot.com", false, "subdomain.blogspot.com"],
   ].freeze
 
+  # rubocop:disable Style/CombinableLoops
   def test_ignore_private
     # test domain and parse
     INCLUDE_PRIVATE_CASES.each do |given, ignore_private, expected|
-      assert_equal expected, PublicSuffix.domain(given, ignore_private: ignore_private)
+      if expected.nil?
+        assert_nil PublicSuffix.domain(given, ignore_private: ignore_private)
+      else
+        assert_equal expected, PublicSuffix.domain(given, ignore_private: ignore_private)
+      end
     end
     # test valid?
     INCLUDE_PRIVATE_CASES.each do |given, ignore_private, expected|
       assert_equal !expected.nil?, PublicSuffix.valid?(given, ignore_private: ignore_private)
     end
   end
+  # rubocop:enable Style/CombinableLoops
 
 
   def valid_uri?(name)
     uri = URI.parse(name)
     !uri.host.nil?
-  rescue
+  rescue StandardError
     false
   end
 
   def valid_domain?(name)
     uri = URI.parse(name)
     !uri.host.nil? && uri.scheme.nil?
-  rescue
+  rescue StandardError
     false
   end
 

data/test/benchmarks/bm_find.rb

@@ -0,0 +1,66 @@
+require 'benchmark'
+require_relative "../../lib/public_suffix"
+
+NAME_SHORT  = "example.de"
+NAME_MEDIUM = "www.subdomain.example.de"
+NAME_LONG   = "one.two.three.four.five.example.de"
+NAME_WILD   = "one.two.three.four.five.example.bd"
+NAME_EXCP   = "one.two.three.four.five.www.ck"
+
+IAAA = "www.example.ac"
+IZZZ = "www.example.zone"
+
+PAAA = "one.two.three.four.five.example.beep.pl"
+PZZZ = "one.two.three.four.five.example.now.sh"
+
+JP   = "www.yokoshibahikari.chiba.jp"
+IT   = "www.example.it"
+COM  = "www.example.com"
+
+TIMES = (ARGV.first || 50_000).to_i
+
+# Initialize
+PublicSuffixList = PublicSuffix::List.default
+PublicSuffixList.find("example.com")
+
+Benchmark.bmbm(25) do |x|
+  x.report("NAME_SHORT") do
+    TIMES.times { PublicSuffixList.find(NAME_SHORT) != nil }
+  end
+  x.report("NAME_MEDIUM") do
+    TIMES.times { PublicSuffixList.find(NAME_MEDIUM) != nil }
+  end
+  x.report("NAME_LONG") do
+    TIMES.times { PublicSuffixList.find(NAME_LONG) != nil }
+  end
+  x.report("NAME_WILD") do
+    TIMES.times { PublicSuffixList.find(NAME_WILD) != nil }
+  end
+  x.report("NAME_EXCP") do
+    TIMES.times { PublicSuffixList.find(NAME_EXCP) != nil }
+  end
+
+  x.report("IAAA") do
+    TIMES.times { PublicSuffixList.find(IAAA) != nil }
+  end
+  x.report("IZZZ") do
+    TIMES.times { PublicSuffixList.find(IZZZ) != nil }
+  end
+
+  x.report("PAAA") do
+    TIMES.times { PublicSuffixList.find(PAAA) != nil }
+  end
+  x.report("PZZZ") do
+    TIMES.times { PublicSuffixList.find(PZZZ) != nil }
+  end
+
+  x.report("JP") do
+    TIMES.times { PublicSuffixList.find(JP) != nil }
+  end
+  x.report("IT") do
+    TIMES.times { PublicSuffixList.find(IT) != nil }
+  end
+  x.report("COM") do
+    TIMES.times { PublicSuffixList.find(COM) != nil }
+  end
+end

data/test/benchmarks/bm_find_all.rb

@@ -0,0 +1,102 @@
+require 'benchmark'
+require_relative "../../lib/public_suffix"
+
+NAME_SHORT  = "example.de"
+NAME_MEDIUM = "www.subdomain.example.de"
+NAME_LONG   = "one.two.three.four.five.example.de"
+NAME_WILD   = "one.two.three.four.five.example.bd"
+NAME_EXCP   = "one.two.three.four.five.www.ck"
+
+IAAA = "www.example.ac"
+IZZZ = "www.example.zone"
+
+PAAA = "one.two.three.four.five.example.beep.pl"
+PZZZ = "one.two.three.four.five.example.now.sh"
+
+JP   = "www.yokoshibahikari.chiba.jp"
+IT   = "www.example.it"
+COM  = "www.example.com"
+
+TIMES = (ARGV.first || 50_000).to_i
+
+# Initialize
+PublicSuffixList = PublicSuffix::List.default
+PublicSuffixList.find("example.com")
+
+Benchmark.bmbm(25) do |x|
+  x.report("NAME_SHORT") do
+    TIMES.times { PublicSuffixList.find(NAME_SHORT) != nil }
+  end
+  x.report("NAME_SHORT (noprivate)") do
+    TIMES.times { PublicSuffixList.find(NAME_SHORT, ignore_private: true) != nil }
+  end
+  x.report("NAME_MEDIUM") do
+    TIMES.times { PublicSuffixList.find(NAME_MEDIUM) != nil }
+  end
+  x.report("NAME_MEDIUM (noprivate)") do
+    TIMES.times { PublicSuffixList.find(NAME_MEDIUM, ignore_private: true) != nil }
+  end
+  x.report("NAME_LONG") do
+    TIMES.times { PublicSuffixList.find(NAME_LONG) != nil }
+  end
+  x.report("NAME_LONG (noprivate)") do
+    TIMES.times { PublicSuffixList.find(NAME_LONG, ignore_private: true) != nil }
+  end
+  x.report("NAME_WILD") do
+    TIMES.times { PublicSuffixList.find(NAME_WILD) != nil }
+  end
+  x.report("NAME_WILD (noprivate)") do
+    TIMES.times { PublicSuffixList.find(NAME_WILD, ignore_private: true) != nil }
+  end
+  x.report("NAME_EXCP") do
+    TIMES.times { PublicSuffixList.find(NAME_EXCP) != nil }
+  end
+  x.report("NAME_EXCP (noprivate)") do
+    TIMES.times { PublicSuffixList.find(NAME_EXCP, ignore_private: true) != nil }
+  end
+
+  x.report("IAAA") do
+    TIMES.times { PublicSuffixList.find(IAAA) != nil }
+  end
+  x.report("IAAA (noprivate)") do
+    TIMES.times { PublicSuffixList.find(IAAA, ignore_private: true) != nil }
+  end
+  x.report("IZZZ") do
+    TIMES.times { PublicSuffixList.find(IZZZ) != nil }
+  end
+  x.report("IZZZ (noprivate)") do
+    TIMES.times { PublicSuffixList.find(IZZZ, ignore_private: true) != nil }
+  end
+
+  x.report("PAAA") do
+    TIMES.times { PublicSuffixList.find(PAAA) != nil }
+  end
+  x.report("PAAA (noprivate)") do
+    TIMES.times { PublicSuffixList.find(PAAA, ignore_private: true) != nil }
+  end
+  x.report("PZZZ") do
+    TIMES.times { PublicSuffixList.find(PZZZ) != nil }
+  end
+  x.report("PZZZ (noprivate)") do
+    TIMES.times { PublicSuffixList.find(PZZZ, ignore_private: true) != nil }
+  end
+
+  x.report("JP") do
+    TIMES.times { PublicSuffixList.find(JP) != nil }
+  end
+  x.report("JP (noprivate)") do
+    TIMES.times { PublicSuffixList.find(JP, ignore_private: true) != nil }
+  end
+  x.report("IT") do
+    TIMES.times { PublicSuffixList.find(IT) != nil }
+  end
+  x.report("IT (noprivate)") do
+    TIMES.times { PublicSuffixList.find(IT, ignore_private: true) != nil }
+  end
+  x.report("COM") do
+    TIMES.times { PublicSuffixList.find(COM) != nil }
+  end
+  x.report("COM (noprivate)") do
+    TIMES.times { PublicSuffixList.find(COM, ignore_private: true) != nil }
+  end
+end

data/test/benchmarks/bm_names.rb

@@ -0,0 +1,91 @@
+require 'benchmark/ips'
+
+STRING = "www.subdomain.example.com"
+ARRAY  = %w(
+  com
+  example.com
+  subdomain.example.com
+  www.subdomain.example.com
+)
+
+def tokenizer1(string)
+  parts = string.split(".").reverse!
+  index = 0
+  query = parts[index]
+  names = []
+
+  loop do
+    names << query
+
+    index += 1
+    break if index >= parts.size
+    query = parts[index] + "." + query
+  end
+  names
+end
+
+def tokenizer2(string)
+  parts = string.split(".")
+  index = parts.size - 1
+  query = parts[index]
+  names = []
+
+  loop do
+    names << query
+
+    index -= 1
+    break if index < 0
+    query = parts[index] + "." + query
+  end
+  names
+end
+
+def tokenizer3(string)
+  isx = string.size
+  idx = string.size - 1
+  names = []
+
+  loop do
+    isx = string.rindex(".", isx - 1) || -1
+    names << string[isx + 1, idx - isx]
+
+    break if isx <= 0
+  end
+  names
+end
+
+def tokenizer4(string)
+  isx = string.size
+  idx = string.size - 1
+  names = []
+
+  loop do
+    isx = string.rindex(".", isx - 1) || -1
+    names << string[(isx+1)..idx]
+
+    break if isx <= 0
+  end
+  names
+end
+
+(x = tokenizer1(STRING)) == ARRAY or fail("tokenizer1 failed: #{x.inspect}")
+(x = tokenizer2(STRING)) == ARRAY or fail("tokenizer2 failed: #{x.inspect}")
+(x = tokenizer3(STRING)) == ARRAY or fail("tokenizer3 failed: #{x.inspect}")
+(x = tokenizer4(STRING)) == ARRAY or fail("tokenizer4 failed: #{x.inspect}")
+
+Benchmark.ips do |x|
+  x.report("tokenizer1") do
+    tokenizer1(STRING).is_a?(Array)
+  end
+  x.report("tokenizer2") do
+    tokenizer2(STRING).is_a?(Array)
+  end
+  x.report("tokenizer3") do
+    tokenizer3(STRING).is_a?(Array)
+  end
+  x.report("tokenizer4") do
+    tokenizer4(STRING).is_a?(Array)
+  end
+
+  x.compare!
+end

data/test/benchmarks/bm_select.rb

@@ -0,0 +1,26 @@
+require 'benchmark'
+require_relative "../../lib/public_suffix"
+
+JP   = "www.yokoshibahikari.chiba.jp"
+
+TIMES = (ARGV.first || 50_000).to_i
+
+# Initialize
+class PublicSuffix::List
+  public :select
+end
+PublicSuffixList = PublicSuffix::List.default
+PublicSuffixList.select("example.jp")
+PublicSuffixList.find("example.jp")
+
+Benchmark.bmbm(25) do |x|
+  x.report("JP select") do
+    TIMES.times { PublicSuffixList.select(JP) }
+  end
+  x.report("JP find") do
+    TIMES.times { PublicSuffixList.find(JP) }
+  end
+  # x.report("JP (noprivate)") do
+  #   TIMES.times { PublicSuffixList.find(JP, ignore_private: true) != nil }
+  # end
+end

data/test/benchmarks/bm_select_incremental.rb

@@ -0,0 +1,25 @@
+require 'benchmark'
+require_relative "../../lib/public_suffix"
+
+JP   = "www.yokoshibahikari.chiba.jp"
+
+TIMES = (ARGV.first || 50_000).to_i
+
+# Initialize
+class PublicSuffix::List
+  public :select
+end
+PublicSuffixList = PublicSuffix::List.default
+PublicSuffixList.select("example.jp")
+
+Benchmark.bmbm(25) do |x|
+  x.report("select jp") do
+    TIMES.times { PublicSuffixList.select("jp") }
+  end
+  x.report("select example.jp") do
+    TIMES.times { PublicSuffixList.select("example.jp") }
+  end
+  x.report("select www.example.jp") do
+    TIMES.times { PublicSuffixList.select("www.example.jp") }
+  end
+end