gman 3.1.1 → 4.0.0

data/lib/gman.rb

@@ -46,6 +46,28 @@ class Gman
         $
       }xi
 
+  # Second level .us domains for states and locality
+  # See http://en.wikipedia.org/wiki/.us
+  #
+  # Examples:
+  #  * foo.state.il.us
+  #  * ci.foo.il.us
+  #
+  # Not:
+  #  * state.foo.il.us
+  #  * foo.ci.il.us
+  #  * k12.il.us
+  #  * ci.foo.zx.us
+  LOCALITY_REGEX = %r{
+    (
+      (state|dst|cog)
+    |
+      (ci|town|vil|co)\.[a-z-]+
+    )
+    \.(ak|al|ar|az|ca|co|ct|dc|de|fl|ga|hi|ia|id|il|in|ks|ky|la|ma|md|me|mi|mn|mo|ms|mt|nc|nd|ne|nh|nj|nm|nv|ny|oh|ok|or|pa|ri|sc|sd|tn|tx|um|ut|va|vt|wa|wi|wv|wy)
+    \.us
+     }x
+
   # Map last part of TLD to alpha2 country code
   ALPHA2_MAP = {
     :ac     => 'sh',
@@ -145,6 +167,9 @@ class Gman
     # Ensure non-edu
     return false if Swot::is_academic?(domain)
 
+    # Check for locality by regex
+    return true if locality?
+
     # check using public suffix's standard logic
     rule = Gman.list.find domain
     return true if !rule.nil? && rule.allow?(domain)
@@ -160,6 +185,11 @@ class Gman
     !!(@text =~ EMAIL_REGEX)
   end
 
+  # Is this domain a .us state or locality?
+  def locality?
+    !!(domain =~ LOCALITY_REGEX)
+  end
+
   # Helper function to return the public suffix domain object
   #
   # Supports all domain strings (URLs, emails)
@@ -196,4 +226,5 @@ class Gman
   def inspect
     "#<Gman domain=\"#{domain}\" valid=#{valid?}>"
   end
+
 end

data/lib/gman/version.rb

@@ -1,3 +1,3 @@
 class Gman
-  VERSION = '3.1.1'
+  VERSION = '4.0.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: gman
 version: !ruby/object:Gem::Version
-  version: 3.1.1
+  version: 4.0.0
 platform: ruby
 authors:
 - Ben Balter
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-08-18 00:00:00.000000000 Z
+date: 2014-09-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: public_suffix
@@ -130,21 +130,14 @@ executables:
 extensions: []
 extra_rdoc_files: []
 files:
-- ".gitignore"
 - CONTRIBUTING.md
 - Gemfile
 - LICENSE
 - README.md
-- Rakefile
 - bin/gman_filter
-- gman.gemspec
 - lib/domains.txt
 - lib/gman.rb
 - lib/gman/version.rb
-- script/build
-- script/release
-- test/helper.rb
-- test/test_gman.rb
 homepage: https://github.com/benbalter/gman
 licenses:
 - MIT

data/.gitignore

@@ -1,4 +0,0 @@
-*.gem
-.bundle
-tmp
-Gemfile.lock

data/Rakefile

@@ -1,29 +0,0 @@
-require 'rubygems'
-require 'bundler'
-begin
-  Bundler.setup(:default, :development)
-rescue Bundler::BundlerError => e
-  $stderr.puts e.message
-  $stderr.puts "Run `bundle install` to install missing gems"
-  exit e.status_code
-end
-require 'rake'
-
-require 'rake/testtask'
-Rake::TestTask.new(:test) do |test|
-  test.libs << 'lib' << 'test'
-  test.pattern = 'test/**/test_gman*.rb'
-  test.verbose = true
-end
-
-desc "Open console with gman loaded"
-task :console do
-  exec "irb -r ./lib/gman.rb"
-end
-
-desc "Validate the domain list"
-Rake::TestTask.new(:domains) do |test|
-  test.libs << 'lib' << 'test'
-  test.pattern = 'test/**/test_domains*.rb'
-  test.verbose = true
-end

data/gman.gemspec

@@ -1,43 +0,0 @@
-require File.expand_path("lib/gman/version", File.dirname(__FILE__))
-
-Gem::Specification.new do |s|
-  s.name = "gman"
-  s.summary = "Check if a given domain or email address belong to a governemnt entity"
-  s.description = "A ruby gem to check if the owner of a given email address is working for THE MAN."
-  s.version = Gman::VERSION
-  s.authors = ["Ben Balter"]
-  s.email = "ben.balter@github.com"
-  s.homepage = "https://github.com/benbalter/gman"
-  s.licenses = ["MIT"]
-
-  s.files = [
-    "Gemfile",
-    "README.md",
-    "CONTRIBUTING.md",
-    "LICENSE",
-    "lib/gman.rb",
-    "lib/domains.txt",
-    "lib/gman/version.rb",
-    "script/build",
-    "script/release",
-    "bin/gman_filter",
-    "gman.gemspec",
-    "test/helper.rb",
-    "test/test_gman.rb",
-    "Rakefile",
-    ".gitignore"
-  ]
-
-  s.executables << "gman_filter"
-
-  s.require_paths = ["lib"]
-  s.add_dependency( "public_suffix", '~> 1.4')
-  s.add_dependency( "swot", '~> 0.3.1' )
-  s.add_dependency( "addressable", '~> 2.3' )
-  s.add_dependency( "iso_country_codes", "~> 0.4" )
-
-  s.add_development_dependency( "rake" )
-  s.add_development_dependency( "shoulda" )
-  s.add_development_dependency( "rdoc" )
-  s.add_development_dependency( "bundler" )
-end

data/script/build

@@ -1,27 +0,0 @@
-#!/usr/bin/env ruby
-# Propagates an initial list of best-guess government domains
-
-require "public_suffix"
-require "yaml"
-
-# https://gist.github.com/benbalter/6147066
-REGEX = /(\.g[ou]{1,2}(v|b|vt)|\.mil|\.gc|\.fed)(\.[a-z]{2})?$/i
-
-YAML_FILE = File.dirname(__FILE__) + "/../lib/domains.yml"
-domains = YAML.load_file YAML_FILE
-domains = [] unless domains
-
-PublicSuffix::List.default.each do |rule|
-  domain = nil
-
-  if rule.parts.length == 1
-    domain = rule.parts.first if ".#{rule.value}" =~ REGEX
-  else
-    domain = rule.parts.pop(2).join(".") if ".#{rule.value}" =~ REGEX
-  end
-
-  domains.push domain unless domain.nil? or domains.include? domain
-end
-
-domains = domains.sort
-File.open(YAML_FILE, 'w+') {|f| f.write(domains.to_yaml)}

data/script/release

@@ -1,38 +0,0 @@
-#!/bin/sh
-# Tag and push a release.
-
-set -e
-
-# Make sure we're in the project root.
-
-cd $(dirname "$0")/..
-
-# Build a new gem archive.
-
-rm -rf gman-*.gem
-gem build -q gman.gemspec
-
-# Make sure we're on the master branch.
-
-(git branch | grep -q '* master') || {
-  echo "Only release from the master branch."
-  exit 1
-}
-
-# Figure out what version we're releasing.
-
-tag=v`ls gman-*.gem | sed 's/^gman-\(.*\)\.gem$/\1/'`
-
-# Make sure we haven't released this version before.
-
-git fetch -t origin
-
-(git tag -l | grep -q "$tag") && {
-  echo "Whoops, there's already a '${tag}' tag."
-  exit 1
-}
-
-# Tag it and bag it.
-
-gem push gman-*.gem && git tag "$tag" &&
-  git push origin master && git push origin "$tag"

data/test/helper.rb

@@ -1,21 +0,0 @@
-require 'rubygems'
-require 'bundler'
-require 'minitest/autorun'
-
-begin
-  Bundler.setup(:default, :development)
-rescue Bundler::BundlerError => e
-  $stderr.puts e.message
-  $stderr.puts "Run `bundle install` to install missing gems"
-  exit e.status_code
-end
-
-require 'shoulda'
-
-$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
-$LOAD_PATH.unshift(File.dirname(__FILE__))
-require 'gman'
-require 'net/dns'
-require 'net/dns/resolver'
-require './lib/gman/parser'
-

data/test/test_gman.rb

@@ -1,94 +0,0 @@
-require File.join(File.dirname(__FILE__), 'helper')
-
-VALID = [  "foo.gov",
-            "http://foo.mil",
-            "foo@bar.gc.ca",
-            "foo.gov.au",
-            "https://www.foo.gouv.fr",
-            "foo@ci.champaign.il.us",
-            "foo.bar.baz.gov.au",
-            "foo@bar.gov.uk",
-            ".gov",
-            "foo.fed.us",
-        ]
-
-INVALID = [ "foo.bar.com",
-            "bar@foo.biz",
-            "http://www.foo.biz",
-            "foo.uk",
-            "gov",
-            "foo@k12.champaign.il.us",
-            "foo@kii.gov.by",
-            "foo",
-            "",
-            nil,
-            " ",
-          ]
-
-class TestGman < Minitest::Test
-
-  should "recognize government email addresses and domains" do
-    VALID.each do |test|
-      assert_equal true, Gman::valid?(test), "#{test} should be detected as a valid government domain"
-    end
-  end
-
-  should "not recognize non-government email addresses and domains" do
-    INVALID.each do |test|
-      assert_equal false, Gman::valid?(test), "#{test} should be detected as an invalid government domain"
-    end
-  end
-
-  should "not allow educational domains" do
-    assert_equal false, Gman::valid?("foo@gwu.edu")
-  end
-
-  should "properly parse domains from strings" do
-    assert_equal "github.gov", Gman.new("foo@github.gov").domain
-    assert_equal "foo.github.gov", Gman::new("foo.github.gov").domain
-    assert_equal "github.gov", Gman::new("http://github.gov").domain
-    assert_equal "github.gov", Gman::new("https://github.gov").domain
-    assert_equal ".gov", Gman::new(".gov").domain
-    assert_equal nil, Gman.new("foo").domain
-  end
-
-  should "not err out on invalid domains" do
-    assert_equal false, Gman.valid?("foo@gov.invalid")
-    assert_equal "gov.invalid", Gman.new("foo@gov.invalid").domain
-    assert_equal nil, Gman.new("foo@gov.invalid").domain_parts
-  end
-
-  should "return public suffix domain" do
-    assert_equal PublicSuffix::Domain, Gman.new("whitehouse.gov").domain_parts.class
-    assert_equal NilClass, Gman.new("foo.invalid").domain_parts.class
-  end
-
-  should "parse domain parts" do
-    assert_equal "gov", Gman.new("foo@bar.gov").domain_parts.tld
-    assert_equal "bar", Gman.new("foo.bar.gov").domain_parts.sld
-    assert_equal "bar", Gman.new("https://foo.bar.gov").domain_parts.sld
-    assert_equal "bar.gov", Gman.new("foo@bar.gov").domain_parts.domain
-  end
-
-  should "not err out on invalid hosts" do
-    assert_equal nil, Gman.new("</@foo.com").domain
-  end
-
-  should "returns the path to domains.txt" do
-    assert_equal true, File.exists?(Gman.list_path)
-  end
-
-  should "parse the alpha2" do
-    assert_equal "us", Gman.new("whitehouse.gov").alpha2
-    assert_equal "us", Gman.new("army.mil").alpha2
-    assert_equal "gb", Gman.new("foo.gov.uk").alpha2
-    assert_equal "ca", Gman.new("gov.ca").alpha2
-  end
-
-  should "determine a domain's country" do
-    assert_equal "United States", Gman.new("whitehouse.gov").country.name
-    assert_equal "United States", Gman.new("army.mil").country.name
-    assert_equal "United Kingdom", Gman.new("foo.gov.uk").country.name
-    assert_equal "Canada", Gman.new("foo.gc.ca").country.name
-  end
-end