profanalyzer 1.0.0 → 1.2.0

data/History.txt

@@ -1,3 +1,8 @@
+=== 1.2.0 / 2011-08-11
+
+* Added #flagged_words to find the discovered offending words. (Thanks nilmethod/Chris Hobbs!)
+* You can now instantiate Profanalyzer to have multiple analyzers with potentially different settings.
+
 === 0.2.1 / 2009-03-25
 
 * Fixed some wordlist errors.

data/Manifest.txt

@@ -1,6 +1,6 @@
 History.txt
 Manifest.txt
-README.markdown
+README.md
 Rakefile
 config/list.yml
 lib/profanalyzer.rb

data/{README.markdown → README.md}

@@ -8,7 +8,7 @@ Profanalyzer has one purpose: analyze a block of text for profanity. It is able
 
 What sets it slightly apart from other filters is that it classifies each blocked word as "profane", "racist", or "sexual" - although right now, each word is considered "profane". It also rates each word on a scale from 0-5,  which is based on my subjective opinion, as well as whether the word is commonly used in non-profane situations, such as "ass" in "assess".
 
-The Profanalyzer will default to a tolerance of of 2, which will kick back the arguably non-profane words. It will also test against all words, including racist or sexual words.
+The Profanalyzer will default to a tolerance of of 4, which will kick back the arguably non-profane words. It will also test against all words, including racist or sexual words.
 
 Lastly, it allows for custom substitutions! For example, the filter at the website http://www.fark.com/ turns the word "fuck" into "fark", and "shit" into "shiat". You can specify these if you want.
 
@@ -45,6 +45,19 @@ Or do specific checking:
     Profanalyzer.profane? "mick" #==> false
     Profanalyzer.profane? "vagina" #==> true
 
+You can obtain a list of the words which fell afoul of profanity checking:
+
+    Profanalyzer.flagged_words("shit damn foo") #==> ["shit", "damn"] 
+    Profanalyzer.flagged_words("profanalyzer is rad!") #==> [] 
+    
+    # With custom settings
+    Profanalyzer.check_all = false
+    Profanalyzer.check_racist = false
+    Profanalyzer.flagged_words("you're a mick") #==> []
+    
+    # You can pass options to the method itself:
+    Profanalyzer.flagged_words("you're a mick", :racist => false) #==> []
+
 Lastly, you can add custom substitutions:
 
     Profanalyzer.substitute("shit","shiat")
@@ -53,6 +66,19 @@ Lastly, you can add custom substitutions:
     Profanalyzer.substitute(:fuck => :fark)
     Profanalyzer.filter("fuck") #==> "fark"
 
+## Non-Global-State use
+
+If you want to not just use global state everywhere, perhaps because you
+need different profanity settings in different contexts, simply create an
+instance of the Profanalyzer class, and use the same methods you were
+using before on the instance:
+
+    analyzer = Profanalyzer.new
+    analyzer.tolerance = 5
+    analyzer.profane? 'hooker' #==> false
+    analyzer.filter 'fuck' #==> '#!$%'
+
+Changing this instance's settings won't affect any other analyzers.
 
 ## Requirements
 
@@ -62,6 +88,8 @@ hoe - a gem for building gems, which I used for profanalyzer.
 
 * Michael Edgar <adgar@carboni.ca>
 * Thomas Hanley <tjhanley.com@gmail.com>
+* Peter Vandenberk <pvandenberk@mac.com>
+* Christopher M. Hobbs <chris@altbit.org> (nilmethod)
 
 ## Installation
 

data/Rakefile

@@ -4,27 +4,10 @@ require 'rubygems'
 require 'hoe'
 require './lib/profanalyzer.rb'
 
-Hoe.new('profanalyzer', Profanalyzer::VERSION) do |p|
-  p.rubyforge_name = 'profanalyzer' # if different than lowercase project name
-  p.developer('Michael J. Edgar', 'edgar@triqweb.com')
+Hoe.spec('profanalyzer') do |p|
+  p.developer('Michael J. Edgar', 'adgar@carboni.ca')
   p.remote_rdoc_dir = ''
   p.summary = "Analyzes a block of text for profanity. It is able to filter profane words as well."
-  desc 'Post your blog announcement to blogger.'
-  task :post_blogger do
-    require 'blogger'
-    p.with_config do |config, path|
-      break unless config['blogs']
-      subject, title, body, urls = p.announcement
-      
-      config['blogs'].each do |site|
-        next unless site['url'] =~ /www\.blogger\.com/
-        acc = Blogger::Account.new(site['user'],site['password'])
-        post = Blogger::Post.new(:title => title, :content => body, :categories => p.blog_categories, :formatter => :rdiscount)
-        acc.post(site['blog_id'], post)
-
-      end
-    end
-  end
   desc 'Pushes rdocs to carbonica'
   task :carbonica => :redocs do
     sh "scp -r doc/ adgar@carboni.ca@carboni.ca:/var/www/html/projects/#{p.name}/"

data/lib/profanalyzer.rb

@@ -98,34 +98,61 @@ require 'yaml'
 # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 class Profanalyzer
   
-  VERSION = "1.0.0"
+  VERSION = "1.2.0"
   
-  @@full_list = YAML::load_file(File.dirname(__FILE__)+"/../config/list.yml")
-  @@racist_list = @@full_list.select {|w| w[:racist]}
-  @@sexual_list = @@full_list.select {|w| w[:sexual]}
+  DEFAULT_TOLERANCE = 4
   
-  @@settings = {:racism => :forbidden, :sexual => :forbidden, :profane => :forbidden, :tolerance => 4, :custom_subs => {}}
+  FULL = YAML::load_file(File.dirname(__FILE__)+"/../config/list.yml")
+  RACIST = FULL.select {|w| w[:racist]}
+  SEXUAL = FULL.select {|w| w[:sexual]}
   
-  def self.forbidden_words_from_settings # :nodoc:
+  DEFAULT_SETTINGS = {:racism => :forbidden,
+                      :sexual => :forbidden,
+                      :profane => :forbidden,
+                      :tolerance => DEFAULT_TOLERANCE,
+                      :custom_subs => {}}
+  
+  def self.singleton_class
+    class << self; self; end
+  end unless defined?(singleton_class)
+
+  def self.forward_to_default(*methods)
+    methods.each do |method|
+      singleton_class.class_eval do
+        define_method method do |*args|
+          DEFAULT_INSTANCE.send(method, *args)
+        end
+      end
+    end
+  end
+
+  def initialize(settings=DEFAULT_SETTINGS)
+    @settings = DEFAULT_SETTINGS
+  end
+
+  DEFAULT_INSTANCE = new
+
+  def forbidden_words_from_settings # :nodoc:
     banned_words = []
     
-    @@full_list.each do |word|
-      banned_words << word[:word] if @@settings[:tolerance] <= word[:badness]
-    end if @@settings[:profane] == :forbidden
+    FULL.each do |word|
+      banned_words << word[:word] if @settings[:tolerance] <= word[:badness]
+    end if @settings[:profane] == :forbidden
     
-    return banned_words if @@settings[:profane] == :forbidden #save some processing
+    return banned_words if @settings[:profane] == :forbidden #save some processing
     
-    @@racist_list.each do |word|
-      banned_words << word[:word] if @@settings[:tolerance] <= word[:badness]
-    end if @@settings[:racism] == :forbidden
+    RACIST.each do |word|
+      banned_words << word[:word] if @settings[:tolerance] <= word[:badness]
+    end if @settings[:racism] == :forbidden
     
-    @@sexual_list.each do |word|
-      banned_words << word[:word] if @@settings[:tolerance] <= word[:badness]
-    end if @@settings[:sexual] == :forbidden
+    SEXUAL.each do |word|
+      banned_words << word[:word] if @settings[:tolerance] <= word[:badness]
+    end if @settings[:sexual] == :forbidden
     banned_words
   end
+  forward_to_default :forbidden_words_from_settings
   
-  def self.update_settings_from_hash(hash)
+  def update_settings_from_hash(hash)
     self.tolerance = hash[:tolerance] if hash.has_key? :tolerance
     self.check_racist = hash[:racist] if hash.has_key? :racist
     self.check_sexual = hash[:sexual] if hash.has_key? :sexual
@@ -137,6 +164,7 @@ class Profanalyzer
       self.check_all = true
     end
   end
+  forward_to_default :update_settings_from_hash
   
   # Decides whether the given string is profane, given Profanalyzer's current
   # settings. Examples:
@@ -157,23 +185,65 @@ class Profanalyzer
   # [:+racist+]  Set to +true+ or +false+ to specify racial slur checking
   # [:+tolerance+] Sets the tolerance. 0-5.
   #
-  def self.profane?(*args)
+  def profane?(*args)
     str = args[0]
     if (args.size > 1 && args[1].is_a?(Hash))
-      oldsettings = @@settings
-      self.update_settings_from_hash args[1]
+      oldsettings = @settings
+      update_settings_from_hash args[1]
     end
-    banned_words = self.forbidden_words_from_settings
+    banned_words = forbidden_words_from_settings
     banned_words.each do |word|
       if str =~ /\b#{word}\b/i
-        @@settings = oldsettings if oldsettings
+        @settings = oldsettings if oldsettings
         return true
       end
     end
-    @@settings = oldsettings if oldsettings
+    @settings = oldsettings if oldsettings
     false
   end
-  
+  forward_to_default :profane?
+ 
+  # Returns an array of words that match the currently set rules against the
+  # provided string.  The array will be empty if no words are matched.
+  #
+  # Example:
+  #    Profanalyzer.flagged_words("shit damn foo") #==> ["shit", "damn"] 
+  #    Profanalyzer.flagged_words("profanalyzer is rad!") #==> [] 
+  #
+  ## With custom settings
+  #    Profanalyzer.check_all = false
+  #    Profanalyzer.check_racist = false
+  #    Profanalyzer.flagged_words("you're a mick") #==> []
+  #
+  # You can pass options to the method itself:
+  #    Profanalyzer.flagged_words("you're a mick", :racist => false) #==> []
+  #
+  # Available options:
+  # 
+  # [:+all+]     Set to +true+ or +false+ to specify checking all words in the blacklist
+  # [:+sexual+]  Set to +true+ or +false+ to specify sexual checking
+  # [:+racist+]  Set to +true+ or +false+ to specify racial slur checking
+  # [:+tolerance+] Sets the tolerance. 0-5.
+  def flagged_words(*args)
+    flagged_words = []
+    str = args[0]
+
+    if (args.size > 1 && args[1].is_a?(Hash))
+      oldsettings = @settings
+      update_settings_from_hash args[1]
+    end
+
+    banned_words = forbidden_words_from_settings
+    banned_words.each do |word|
+      if str =~ /\b#{word}\b/i
+        flagged_words << word
+      end
+    end
+    @settings = oldsettings if oldsettings
+    return flagged_words
+  end
+  forward_to_default :flagged_words
+
   # Filters the provided string using the currently set rules, with #!@$%-like
   # characters substituted in.
   #
@@ -195,16 +265,16 @@ class Profanalyzer
   # [:+racist+]  Set to +true+ or +false+ to specify racial slur checking
   # [:+tolerance+] Sets the tolerance. 0-5.
   #
-  def self.filter(*args)
+  def filter(*args)
     str = args[0]
     if (args.size > 1 && args[1].is_a?(Hash))
-      oldsettings = @@settings
+      oldsettings = @settings
       self.update_settings_from_hash args[1]
     end
     
     retstr = str
     
-    @@settings[:custom_subs].each do |k,v|
+    @settings[:custom_subs].each do |k,v|
       retstr.gsub!(/\b#{k.to_s}\b/i,v.to_s)
     end
     
@@ -213,20 +283,21 @@ class Profanalyzer
       retstr.gsub!(/\b#{word}\b/i,
           "#!$%@&!$%@%@&!$#!$%@&!$%@%@&!#!$%@&!$%@%@&!"[0..(word.length-1)])
     end
-    @@settings = oldsettings if oldsettings
+    @settings = oldsettings if oldsettings
     retstr
   end                 
-  
-  def self.strip(*args)
+  forward_to_default :filter
+
+  def strip(*args)
     str = args[0]
     if (args.size > 1 && args[1].is_a?(Hash))
-      oldsettings = @@settings
+      oldsettings = @settings
       self.update_settings_from_hash args[1]
     end
     
     retstr = str
     
-    @@settings[:custom_subs].each do |k,v|
+    @settings[:custom_subs].each do |k,v|
       retstr.gsub!(/\b#{k.to_s}\b/i,v.to_s)
     end
     
@@ -234,49 +305,59 @@ class Profanalyzer
     banned_words.each do |word|
       retstr.gsub!(/\b#{word}\b/i,"")
     end
-    @@settings = oldsettings if oldsettings
+    @settings = oldsettings if oldsettings
     retstr
   end 
+  forward_to_default :strip
   
   # Sets Profanalyzer's tolerance. Value should be an integer such that 
   # 0 <= T <= 5.
-  def self.tolerance=(new_tol)
-    @@settings[:tolerance] = new_tol
+  def tolerance=(new_tol)
+    @settings[:tolerance] = new_tol
+  end
+  
+  # Returns Profanalyzer's tolerance. Value will be an integer
+  # 0 <= T <= 5.
+  def tolerance
+    @settings[:tolerance]
   end
   
   # Sets Profanalyzer to scan (or not scan) for racist words, based on 
   # the set tolerance.
   # This is set to +true+ by default.
-  def self.check_racist=(check)
-    @@settings[:racism] = (check) ? :forbidden : :ignore
+  def check_racist=(check)
+    @settings[:racism] = (check) ? :forbidden : :ignore
   end
   
   # Sets Profanalyzer to scan (or not scan) for sexual words, based on the set tolerance.
   # This is set to +true+ by default.
-  def self.check_sexual=(check)
-    @@settings[:sexual] = (check) ? :forbidden : :ignore
+  def check_sexual=(check)
+    @settings[:sexual] = (check) ? :forbidden : :ignore
   end
   
   # Sets Profanalyzer to scan (or not scan) for all profane words, based on the set tolerance.
   # This is set to +true+ by default.
-  def self.check_all=(check)
-    @@settings[:profane] = (check) ? :forbidden : :ignore
+  def check_all=(check)
+    @settings[:profane] = (check) ? :forbidden : :ignore
   end
   
   # Sets the list of substitutions to the hash passed in. Substitutions are
   # performed such that +Profanalyzer.filter(key) = value+.
-  def self.subtitutions=(hash)
-    @@settings[:custom_subs] = hash
+  def substitutions=(hash)
+    @settings[:custom_subs] = hash
   end
   
   # Sets a custom substitution for the filter.
   # Can be passed as +substitute("foo","bar")+ or +"foo" => "bar"+
-  def self.substitute(*args)
+  def substitute(*args)
     case args[0]
     when String
-      @@settings[:custom_subs].merge!(args[0] => args[1])
+      @settings[:custom_subs].merge!(args[0] => args[1])
     when Hash
-      @@settings[:custom_subs].merge!(args[0])
+      @settings[:custom_subs].merge!(args[0])
     end
   end
-end
+
+  forward_to_default :tolerance, :tolerance=, :check_racist=, :check_sexual=, :check_all=
+  forward_to_default :substitutions, :substitute
+end

data/test/test_profanalyzer.rb

@@ -3,10 +3,22 @@ require "profanalyzer"
 
 class TestProfanalyzer < Test::Unit::TestCase
   
+  def test_default_tolerance
+    assert_equal Profanalyzer::DEFAULT_TOLERANCE, Profanalyzer.tolerance
+  end
+  
+  def test_profanalyzer_tolerance
+    0.upto(5) do |tolerance|
+      Profanalyzer.tolerance = tolerance # setter
+      assert_equal tolerance, Profanalyzer.tolerance # getter
+    end
+  end
+  
   def test_single_word
     Profanalyzer.tolerance = 0
     Profanalyzer.check_all = true
     assert_equal(true, Profanalyzer.profane?("asshole"))
+    assert_equal(["asshole"], Profanalyzer.flagged_words("asshole"))
   end
   
   def test_single_racist_word
@@ -16,6 +28,8 @@ class TestProfanalyzer < Test::Unit::TestCase
     Profanalyzer.check_racist = true
     assert_equal(true, Profanalyzer.profane?("spic"))
     assert_equal(false, Profanalyzer.profane?("pussy"))
+    assert_equal(["spic"], Profanalyzer.flagged_words("spic"))
+    assert_equal([], Profanalyzer.flagged_words("pussy"))
   end
   
   def test_single_sexual_word
@@ -25,6 +39,8 @@ class TestProfanalyzer < Test::Unit::TestCase
     Profanalyzer.check_sexual = true
     assert_equal(true, Profanalyzer.profane?("vagina"))
     assert_equal(false, Profanalyzer.profane?("nigger"))
+    assert_equal(["vagina"], Profanalyzer.flagged_words("vagina"))
+    assert_equal([], Profanalyzer.flagged_words("nigger"))
   end
   
   def test_tolerance
@@ -32,6 +48,8 @@ class TestProfanalyzer < Test::Unit::TestCase
     Profanalyzer.check_all = true
     assert_equal(false, Profanalyzer.profane?("asskisser")) # badness = 3
     assert_equal(true, Profanalyzer.profane?("fuck"))       # badness = 5
+    assert_equal([], Profanalyzer.flagged_words("asskisser")) # badness = 3
+    assert_equal(["fuck"], Profanalyzer.flagged_words("fuck"))       # badness = 5
   end
   
   def test_sexual_tolerance
@@ -41,6 +59,8 @@ class TestProfanalyzer < Test::Unit::TestCase
     Profanalyzer.check_sexual = true
     assert_equal(false, Profanalyzer.profane?("vagina")) # badness = 3
     assert_equal(true, Profanalyzer.profane?("cunt"))       # badness = 5
+    assert_equal([], Profanalyzer.flagged_words("vagina")) # badness = 3
+    assert_equal(["cunt"], Profanalyzer.flagged_words("cunt"))       # badness = 5
   end
   
   def test_racist_tolerance
@@ -50,6 +70,8 @@ class TestProfanalyzer < Test::Unit::TestCase
     Profanalyzer.check_racist = true
     assert_equal(false, Profanalyzer.profane?("mick")) # badness = 3
     assert_equal(true, Profanalyzer.profane?("nigger"))       # badness = 5
+    assert_equal([], Profanalyzer.flagged_words("mick")) # badness = 3
+    assert_equal(["nigger"], Profanalyzer.flagged_words("nigger"))       # badness = 5
   end
   
   def test_filter
@@ -80,7 +102,7 @@ class TestProfanalyzer < Test::Unit::TestCase
     assert_equal(filtered_string, Profanalyzer.filter(original_string))
   end
   
-  def test_subtitutions
+  def test_substitutions
     Profanalyzer.substitute("shit","shiat")
     assert_equal("shiat", Profanalyzer.filter("shit"))
     
@@ -91,4 +113,10 @@ class TestProfanalyzer < Test::Unit::TestCase
     assert_equal("fark", Profanalyzer.filter("fuck"))
   end
 
+  def test_multiple_matches_in_flagged_words
+    Profanalyzer.tolerance = 0
+    Profanalyzer.check_all = true
+    assert_equal(["shit", "mick", "cocksucking"], Profanalyzer.flagged_words("You're a cocksucking piece of shit, you mick."))    
+  end
+
 end

data/test/test_profanalyzer_advanced.rb

@@ -5,31 +5,42 @@ class TestProfanalyzer < Test::Unit::TestCase
   
   def test_single_word_advanced
     assert_equal(true, Profanalyzer.profane?("asshole", :tolerance => 0, :all => true))
+    assert_equal(["asshole"], Profanalyzer.flagged_words("asshole", :tolerance => 0, :all => true))
   end
   
   def test_single_racist_word_advanced
     assert_equal(true, Profanalyzer.profane?("spic", :tolerance => 0, :racist => true, :sexual => false))
     assert_equal(false, Profanalyzer.profane?("pussy", :tolerance => 0, :racist => true, :sexual => false))
+    assert_equal(["spic"], Profanalyzer.flagged_words("spic", :tolerance => 0, :racist => true, :sexual => false))
+    assert_equal([], Profanalyzer.flagged_words("pussy", :tolerance => 0, :racist => true, :sexual => false))
   end
   
   def test_single_sexual_word_advanced
     assert_equal(true, Profanalyzer.profane?("vagina", :tolerance => 0, :racist => false, :sexual => true))
     assert_equal(false, Profanalyzer.profane?("nigger", :tolerance => 0, :racist => false, :sexual => true))
+    assert_equal(["vagina"], Profanalyzer.flagged_words("vagina", :tolerance => 0, :racist => false, :sexual => true))
+    assert_equal([], Profanalyzer.flagged_words("nigger", :tolerance => 0, :racist => false, :sexual => true))
   end
   
   def test_tolerance_advanced
     assert_equal(false, Profanalyzer.profane?("asskisser", :tolerance => 4, :all => true)) # badness = 3
     assert_equal(true, Profanalyzer.profane?("fuck", :tolerance => 4, :all => true))       # badness = 5
+    assert_equal([], Profanalyzer.flagged_words("asskisser", :tolerance => 4, :all => true)) # badness = 3
+    assert_equal(["fuck"], Profanalyzer.flagged_words("fuck", :tolerance => 4, :all => true))       # badness = 5
   end
   
   def test_sexual_tolerance_advanced
     assert_equal(false, Profanalyzer.profane?("vagina", :tolerance => 4, :racist => false, :sexual => true)) # badness = 3
     assert_equal(true, Profanalyzer.profane?("cunt", :tolerance => 4, :racist => false, :sexual => true))       # badness = 5
+    assert_equal([], Profanalyzer.flagged_words("vagina", :tolerance => 4, :racist => false, :sexual => true)) # badness = 3
+    assert_equal(["cunt"], Profanalyzer.flagged_words("cunt", :tolerance => 4, :racist => false, :sexual => true))       # badness = 5
   end
   
   def test_racist_tolerance_advanced
     assert_equal(false, Profanalyzer.profane?("mick", :tolerance => 4, :racist => true, :sexual => false)) # badness = 3
     assert_equal(true, Profanalyzer.profane?("nigger", :tolerance => 4, :racist => true, :sexual => false))       # badness = 5
+    assert_equal([], Profanalyzer.flagged_words("mick", :tolerance => 4, :racist => true, :sexual => false)) # badness = 3
+    assert_equal(["nigger"], Profanalyzer.flagged_words("nigger", :tolerance => 4, :racist => true, :sexual => false))       # badness = 5
   end
   
   def test_filter_advanced

metadata

@@ -4,9 +4,9 @@ version: !ruby/object:Gem::Version
   prerelease: false
   segments: 
   - 1
+  - 2
   - 0
-  - 0
-  version: 1.0.0
+  version: 1.2.0
 platform: ruby
 authors: 
 - Michael J. Edgar
@@ -14,7 +14,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-04-06 00:00:00 -04:00
+date: 2011-08-11 00:00:00 -04:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -27,8 +27,8 @@ dependencies:
         segments: 
         - 2
         - 0
-        - 2
-        version: 2.0.2
+        - 4
+        version: 2.0.4
   type: :development
   version_requirements: *id001
 - !ruby/object:Gem::Dependency 
@@ -40,9 +40,9 @@ dependencies:
       - !ruby/object:Gem::Version 
         segments: 
         - 0
-        - 2
+        - 6
         - 1
-        version: 0.2.1
+        version: 0.6.1
   type: :development
   version_requirements: *id002
 - !ruby/object:Gem::Dependency 
@@ -61,7 +61,7 @@ dependencies:
   version_requirements: *id003
 description: ""
 email: 
-- edgar@triqweb.com
+- adgar@carboni.ca
 executables: []
 
 extensions: []
@@ -72,7 +72,7 @@ extra_rdoc_files:
 files: 
 - History.txt
 - Manifest.txt
-- README.markdown
+- README.md
 - Rakefile
 - config/list.yml
 - lib/profanalyzer.rb