despamilator 0.8 → 1.0

data/lib/despamilator/filter/ip_address_url.rb

@@ -1,11 +1,21 @@
-def name
-  'IP Address URL'
-end
+require 'despamilator/filter_base'
 
-def description
-  'Detects IP address URLs'
-end
+module DespamilatorFilter
+
+  class IPAddressURL < Despamilator::FilterBase
+
+    def name
+      'IP Address URL'
+    end
+
+    def description
+      'Detects IP address URLs'
+    end
+
+    def parse text
+      self.append_score = 0.5 if text.downcase.scan(/http:\/\/\d+\.\d+\.\d+\.\d+/).length > 0
+    end
+
+  end
 
-def parse
-  self.append_score = 0.5 if (self.text.downcase.scan(/http:\/\/\d+\.\d+\.\d+\.\d+/).length > 0)
 end

data/lib/despamilator/filter/long_words.rb

@@ -1,13 +1,23 @@
-def name
-  'Long Words'
-end
+require 'despamilator/filter_base'
 
-def description
-  'Detects long and unbroken strings'
-end
+module DespamilatorFilter
+
+  class LongWords < Despamilator::FilterBase
+
+    def name
+      'Long Words'
+    end
+
+    def description
+      'Detects long and unbroken strings'
+    end
+
+    def parse text
+      text.split(/\W+/).each do |word|
+        self.append_score = 0.1 if word.length > 20
+      end
+    end
 
-def parse
-  self.text.split(/\W+/).each do |word|
-    self.append_score = 0.1 if word.length > 20
   end
-end
+
+end

data/lib/despamilator/filter/naughty_q.rb

@@ -1,21 +1,31 @@
-def name
-  'Naughty Q'
-end
+require 'despamilator/filter_base'
 
-def description
-  'Detects possible misuse of the letter Q (English language)'
-end
+module DespamilatorFilter
 
-def parse
-  post_matches = self.text.downcase.scan(/q(\w|\d)/)
-  pre_matches = self.text.downcase.scan(/(\w|\d)q/)
+  class NaughtyQ < Despamilator::FilterBase
 
-  matches = post_matches + pre_matches
+    def name
+      'Naughty Q'
+    end
 
-  return unless matches
+    def description
+      'Detects possible misuse of the letter Q (English language)'
+    end
+
+    def parse text
+      post_matches = text.downcase.scan(/q(\w|\d)/)
+      pre_matches = text.downcase.scan(/(\w|\d)q/)
+
+      matches = post_matches + pre_matches
+
+      return unless matches
+
+      matches.each do |match|
+        match = match.first
+        self.append_score = 0.2 unless match == 'u' or match == 'a' or match == 'k'
+      end
+    end
 
-  matches.each do |match|
-    match = match.first
-    self.append_score = 0.2 unless match == 'u' or match == 'a' or match == 'k'
   end
+
 end

data/lib/despamilator/filter/naughty_words.rb

@@ -1,21 +1,27 @@
-def name
-  'Naughty Words'
-end
+require 'despamilator/filter_base'
 
-def description
-  'Detects cheeky words'
-end
+module DespamilatorFilter
 
-def parse
-  text = self.text.downcase
+  class NaughtyWords < Despamilator::FilterBase
 
-  naughty_words.each do |word|
-    self.append_score = 0.1 if text =~ /\b#{word}\b/
-  end
-end
+    def name
+      'Naughty Words'
+    end
+
+    def description
+      'Detects cheeky words'
+    end
 
-def naughty_words
-  %w{
+    def parse text
+      text.downcase!
+
+      naughty_words.each do |word|
+        self.append_score = 0.1 if text =~ /\b#{word}\b/
+      end
+    end
+
+    def naughty_words
+      %w{
     penis
     viagra
     bondage
@@ -33,5 +39,8 @@ def naughty_words
     cock
     pussy
     clit
-  }
-end
+   }
+    end
+
+  end
+end

data/lib/despamilator/filter/numbers_and_words.rb

@@ -1,33 +1,43 @@
-def parse
-  string = self.text.downcase
-
-  # strip out "good numbers"
-  string.gsub!(/h[1-6]/, '')
-  string.gsub!(/(^|\b)\d+($|\b)/, '')
-  string.gsub!(/(^|\b)\d+(,|\.)\d+($|\b)/, '')
-  string.gsub!(/(^|\b)\d+(st|nd|rd|th)($|\b)/, '')
-
-  [
-          /\w\d+/,
-          /\d+\w/,
-          /\d+($|\b)/
-  ].each do |regexp|
-    matches = string.scan(regexp)
-
-    next if matches.empty?
-
-    matches.each do |to_remove|
-      to_remove = to_remove.to_s
-      string.sub!(to_remove, '') unless to_remove.empty?
-      self.append_score = 0.1
+require 'despamilator/filter_base'
+
+module DespamilatorFilter
+
+  class NumbersAndWords < Despamilator::FilterBase
+
+    def parse text
+      text.downcase!
+
+      # strip out "good numbers"
+      text.gsub!(/h[1-6]/, '')
+      text.gsub!(/(^|\b)\d+($|\b)/, '')
+      text.gsub!(/(^|\b)\d+(,|\.)\d+($|\b)/, '')
+      text.gsub!(/(^|\b)\d+(st|nd|rd|th)($|\b)/, '')
+
+      [
+              /\w\d+/,
+              /\d+\w/,
+              /\d+($|\b)/
+      ].each do |regexp|
+        matches = text.scan(regexp)
+
+        next if matches.empty?
+
+        matches.each do |to_remove|
+          to_remove = to_remove.to_s
+          text.sub!(to_remove, '') unless to_remove.empty?
+          self.append_score = 0.1
+        end
+      end
+    end
+
+    def name
+      'Numbers next to words'
     end
-  end
-end
 
-def name
-  'Numbers next to words'
-end
+    def description
+      'Detects unusual number/word combinations'
+    end
+
+  end
 
-def description
-  'Detects unusual number/word combinations'
 end

data/lib/despamilator/filter/script_tag.rb

@@ -1,13 +1,21 @@
-def parse
-  if self.text.downcase.match(/<\/?script(>|\s+|\n|\r)/)
-    self.append_score = 1
-  end
-end
+require 'despamilator/filter_base'
 
-def name
-  'Detects script tags in text'
-end
+module DespamilatorFilter
+
+  class ScriptTag < Despamilator::FilterBase
+
+    def parse text
+      self.append_score = 1 if text.downcase.match(/<\/?script(>|\s+|\n|\r)/)
+    end
+
+    def name
+      'Script tag'
+    end
+
+    def description
+      'Searches for variations for the HTML script tag'
+    end
+
+  end
 
-def description
-  'Searches for variations for the HTML script tag'
 end

data/lib/despamilator/filter/shouting.rb

@@ -1,21 +1,31 @@
-def name
-  'Shouting'
-end
+require 'despamilator/filter_base'
 
-def description
-  'Detects and scores shouting (all caps)'
-end
+module DespamilatorFilter
+
+  class Shouting < Despamilator::FilterBase
+
+    def name
+      'Shouting'
+    end
 
-def parse
-  # strip HTML
-  text = self.text.gsub(/<\/?[^>]*>/, "")
+    def description
+      'Detects and scores shouting (all caps)'
+    end
 
-  return if text.length < 20
+    def parse text
+      # strip HTML
+      text.gsub!(/<\/?[^>]*>/, "")
 
-  uppercased = text.scan(/[A-Z]/).length
-  lowercased = text.scan(/[a-z]/).length
+      return if text.length < 20
+
+      uppercased = text.scan(/[A-Z]/).length
+      lowercased = text.scan(/[a-z]/).length
+
+      if uppercased > 0
+        self.append_score = (uppercased.to_f / (uppercased + lowercased)) * 0.5
+      end
+    end
 
-  if uppercased > 0
-    self.append_score = (uppercased.to_f / (uppercased + lowercased)) * 0.5
   end
-end
+
+end

data/lib/despamilator/filter/square_brackets.rb

@@ -1,13 +1,23 @@
-def name
-  'Square Brackets'
-end
+require 'despamilator/filter_base'
 
-def description
-  'Detects each square bracket in a string'
-end
+module DespamilatorFilter
+
+  class SquareBrackets < Despamilator::FilterBase
+
+    def name
+      'Square Brackets'
+    end
+
+    def description
+      'Detects each square bracket in a string'
+    end
+
+    def parse text
+      text.downcase.scan(/(\[|\])/).each do |match|
+        self.append_score = 0.05
+      end
+    end
 
-def parse
-  self.text.downcase.scan(/(\[|\])/).each do |match|
-    self.append_score = 0.05
   end
+
 end

data/lib/despamilator/filter/urls.rb

@@ -1,11 +1,27 @@
-def name
-  'URLs'
-end
+require 'despamilator/filter_base'
 
-def description
-  'Detects each url in a string'
-end
+module DespamilatorFilter
+
+  class URLs < Despamilator::FilterBase
+
+    def name
+      'URLs'
+    end
+
+    def description
+      'Detects each url in a string'
+    end
+
+    def parse text
+      text.downcase!
+
+      text.gsub!(/http:\/\/\d+\.\d+\.\d+\.\d+/, '')
+      
+      1.upto(text.scan(/http:\/\//).length) do
+        self.append_score = 0.20
+      end
+    end
+
+  end
 
-def parse
-  self.append_score = 0.20 * self.text.downcase.scan(/http:\/\//).length
 end

data/lib/despamilator/filter_base.rb

@@ -1,27 +1,72 @@
 class Despamilator
+
+#This class is the base class of all the despamilator filters.
+#
+#== EXAMPLE:
+#
+#This example is to detect the letter "a". Put the code in
+#lib/despamilator/filter/detect_letter_a.rb:
+#
+#  require 'despamilator/filter_base'
+#
+#  module DespamilatorFilter
+#
+#    class DetectLetterA < Despamilator::FilterBase
+#
+#      def name
+#        'Detecting the letter A'
+#      end
+#
+#      def description
+#        'Detects the letter "a" in a string for no reason other than a demo'
+#      end
+#
+#      def parse text
+#        if text.downcase.scan(/a/)
+#        # add 0.1 to the score of the text
+#        self.append_score = 0.1
+#      end
+#    end
+#  end
+
   class FilterBase
-    attr_accessor :text, :score, :filename, :matches
+    attr_accessor :text, :score, :matches
+
+    # Constructor for the class.
 
-    def initialize text, filename
+    def initialize
       @matches = 0
-      @filename = filename
       @score = 0
-      @text = text
       @matched = false
     end
 
+    # Returns the score the filter instance has calculated.
+
+    def score
+      # workaround for a bug in ruby 1.9.2's floats
+      sprintf("%.3f", @score).to_f
+    end
+
+    # The nice description of the filter. Usually no more than a sentence.
+
     def description
-      raise "No description defined in #{filename}"
+      raise "No description defined for #{self.class}"
     end
 
-    def parse blah
-      raise "No parser defined in #{filename}"
+    # This method parses some text. The score is assigned to the same instance.
+
+    def parse text
+      raise "No parse defined for #{self.class}"
     end
 
+    # The one or two word name for the filter.
+
     def name
-      raise "No name defined in #{filename}"
+      raise "No name defined for #{self.class}"
     end
-    
+
+    # Boolean. Whether or not the filter matched anything.
+
     def matched?
       @score > 0
     end